(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Measurement of capillary rise under field conditions and related soil properties."

■♦■ 



Agriculture 
Canada 



I dft* Agriculture 



Canada 

Research Direction generale 
Branch de la recherche 



Contribution 1 983-1 9E 



Ottawa KlA e o^ ad,erin ^eragriculture 




J Measurement of capillary rise 
under field conditions and 
related soil properties 




630.72 

C759 

C 83-19E 

OOAg 









!>' 



Canada 



The map on the cover has dots representing 
Agriculture Canada research establishments. 






Measurement of capillary rise 
under field conditions and 
related soil properties 



D. H. WEBSTER > 
Research Station 
Kentville, Nova Scotia 

G.C.TOPP h° " r) ' 

Land Resource Research Institute 
Ottawa, Ontario 

Kentville Research Station Technical Bulletin No. 3 



Research Branch 
Agriculture Canada 
1983 



Copies of this publication are available from: 

Dr. G. M. Weaver 

Director 

Research Station 

Research Branch, Agriculture Canada 

Kentville, Nova Scotia 

B4N 1J5 

Produced by Research Program Service 

©Minister of Supply and Services Canada 1983 



SUMMARY 



Capil 

exper 

herba 

unusu 

of ri 

r e m i n 

pi ant 

Hydra 

sites 

si tes 

the r 

s i m i 1 

woul d 

exper 

facto 

table 

the s 

Corre 

prope 

af f ec 

bulk 

(for 

water 

domi n 

from 

(appr 

for p 



1 ary ri se , 
imental pe 
ceous cul t 
al ly deep 
se, no mea 
der that c 

growth on 
u 1 i c prope 

fell well 

that repr 
egi on (san 
ar water t 

probably 
imental or 
r of 2) wi 

depth and 
econd site 
1 a t i o n s be 
rt i es sugg 
ted by the 
density wa 
a gi ven f 1 

tabl e dep 
ant textur 
water tabl 
oachi ng 2 
1 ant growt 



meas 
ri ods 
u res 
in 1 9 
su rem 
a p i 1 1 
ly if 
r t i e s 

with 
esent 
d, lo 
able 
equal 
chard 
th ri 

soi 1 

were 
tween 
est t 

high 
s ass 
ux ) a 
th. 
al co 
es wi 
g/cm 3 
h und 



u red a 

suppl 

during 

78 and 

ent of 

ary ri 

water 

of co 

i n the 

ed the 

amy sa 

depths 

or ex 

. Obs 

se cal 

moi st 

more 

hydra 

hat ca 

bulk 
oci ate 
nd a g 
Percen 
nt rol s 
thin s 
) i s n 
er fie 



t two s 
ied fro 

these 
, becau 

c a p i 1 1 
s e is a 

table 
re samp 

range 

range 
nd, san 
, ri se 
ceed th 
erved r 
cul ated 
u re ten 
v a r i a b 1 
ul ic pr 
pi 1 1 ary 
den si ty 
d with 
reater 
t sand 

of ri s 
ubsoi 1 s 
ot like 
1 d cond 



i tes 
m 45 
peri o 
se of 
ary r 

rel i 
depth 
les f 
of sa 
of te 
dy lo 
in ma 
e ri s 
i se a 

from 
si on 
e and 
opert 

ri se 

of t 
a nee 
sensi 
and f 
e par 

of u 
ly to 
i t i o n 



l n o 
to 7 
ds. 

the 
i se 
able 

is 
rom 
mpl e 
xtu r 
am a 
ny s 
e th 
gree 

cor 
on o 

agr 
i es 

par 
hese 
d fo 
t i v i 
i nen 
amet 
nusu 

sup 
s. 



ne ore 
9% of 

The w 

conse 
was po 

sourc 
rel at i 
the ca 
s from 
e of o 
nd loa 
oils o 
at was 
d cl os 
e prop 
n e sit 
eement 
and ot 
ameter 

subso 
r shal 
ty of 
ess of 
ers. C 
ally h 
ply ap 



hard 
the w 
ater 
quent 
ssi bl 
e of 
vely 
pi 1 1 a 

ei gh 
rchar 
m) . 
f the 

obse 

ely ( 
e r t i e 
e but 

was 
her s 
s wer 
i 1 s . 
1 ower 
flux 

sand 
a p i 1 1 
i gh b 
preci 



over t 
ater u 
table 

1 ow r 
e i n 1 
water 
stabl e 
ry r i s 
t othe 
d soil 
Thus , 

regi o 
rved i 
within 
s , wat 

cores 
less g 
oil 
e adve 

Incre 

water 
to cha 

were 
ary ri 
ul k de 
able w 



hree 
sed by 
was 
ates 
978; a 
for 

• 

e 

r 

s i n 

given 

n, 

n the 

a 
er 

from 
ood. 

rsely 
ase i n 
tabl e 
nge in 
the 
se 

n s i t y 
ater 



m 



RESUME 



L 'asce 
au cou 
7 9% de 
temp s . 
la v i t 
i mposs 
que 1 ' 
pi ante 
est re 
ca rott 
compa r 
en hu i 
de tex 
1 oameu 
ph reat 
de nom 
s u p e r i 
experi 
e t r o i t 
prop ri 
la ten 
obse r v 
Si 1 'o 
hy d ra u 
1 'asce 
de ces 
as soci 
flux d 
change 
et la 
des pa 
cap i 1 1 
sou s -s 
de v ra i 
assure 



ns i o 

rs d 
1 'e 
En 
esse 
i ble 
asce 
s en 
1 ati 
es d 
a i en 
t au 
t u re 
x, 1 
i que 
breu 
eu re 
ment 
emen 
et es 
s i on 
ees 
n en 
1 i qu 
ns i o 
sou 
ee a 
onne 
ment 
fine 
rame 
a ri t 
ol s 
t pa 
r la 



n cap 

e tro 

au ut 

1978 

d'as 

d'en 

ns i on 

croi 

vemen 

e sol 

t bie 

t res 

s des 

oam s 

s de 

x sol 

a ce 

al . 

t (a 

de 1 

de 1 

e t a i t 

juge 

es et 

n cap 

s -sol 

u bes 

) et 

s de 

s se d 

t res 

e des 

a for 

s su f 

croi 



i 1 1 a i r 

is per 

i 1 i see 

, 1 e n 

censi o 

fai re 

c a p i 1 

ssance 

t stab 

prel e 

nice 

endroi 

sols 
abl eux 
p rof on 
s de 1 
lie qu 
En un 
un f ac 
a ca ro 
1 eau d 
plus 
d ' ap r 
d 'aut 
i 1 1 a i r 
s. L' 
o i n d ' 
a une 
p rof on 
u sabl 
de l'a 
nappe 
te den 
fire a 
ssance 



e , m 
i ode 

par 
i vea 
n a 

la 
lair 

que 
le. 
vees 
1 les 
ts , 
des 

et 
deu r 
a re 
i a 
end r 
teu r 
tte , 
u so 
va ri 
es 1 
res 
e f u 
augm 
u ne 
plus 
deu r 
e et 
seen 
s ph 
site 

en 

des 



esu r 

s ex 

1 es 

u ph 
ete 
mesu 
e n' 

si 

Les 

des 

de 
echa 
verg 
1 oam 

sem 
gion 
ete 
oi t , 

de 

la 
1 , m 
able 
es c 
prop 
rent 
enta 
napp 

gra 

de 
a i en 
si on 
reat 

app 
f ou r 

p 1 a 



ee e 

peri 

cul 

reat 

faib 

re c 

est 

1 a p 

p ro 

si t 

1 'ev 

nti 1 

ers 

). 
blab 

ser 
obse 

1 'a 
2) a 
prof 
a i s 

et 

rre 
ri et 

per 
t i on 
e ph 
nde 

1 a n 
t le 

cap 
i que 
a ren 
n i r 
ntes 



n de 
ment 
tu re 
i que 
le d 
ette 
une 
rof o 
p ri e 
es d 
enta 
1 ons 
de 1 
Par 
le, 
ai t 
rvee 
seen 

eel 
onde 
au d 
1 a c 
1 ati 
es d 
tu rb 

de 
reat 
sens 
appe 
s pr 
ilia 
s qu 
te ( 
une 

dan 



ux e 
al es 
s he 

eta 
e so 

ann 
sour 
ndeu 
tes 
' asc 
il d 

qu i 
a re 
cons 
1 ' as 
prob 

dan 
s i on 
1 e c 
ur d 
eux i 
orre 
ons 
u so 
es p 
la d 
i que 
ibi 1 

ph r 
i nc i 
ire. 
i se 
appr 
quan 
s 1 e 



n d r o i t 
, f ou r 
rbacee 
n t i n h 
rte qu 
ee -1 a 
ce d ' e 
r de 1 
hy d rau 
ension 
1 echan 

rep re 
gi on ( 
equent 
censi o 
abl erne 
s 1 e v 

obser 
a 1 eu 1 e 
elan 
erne en 
sponda 
ent re 
1, les 
ar la 
ens i te 

m o i n s 

i t e d u 

e a t i q u 

paux c 

L'ea 

trou v 
ochant 
t i t e a 
s cond 



s da 
ni ss 
s du 
abi t 
'il 
- ce 
au s 
a na 
1 i qu 
cap 
till 
sent 
sabl 
, po 
n ca 
nt e 
erge 
vee 
e d' 
appe 
d roi 
nee 
1 es 

pa r 
dens 
app 
p ro 
flu 
e . 

ont r 
u mo 
ent 

2 g 
pp re 
i t i o 



ns u 
ai t 
rant 
uell 
a et 

qu i 
u re 
ppe 
es d 
ilia 
ons 
ai en 
e, s 
ur d 
pi 1 1 
gale 
r 

com 
ap re 

phr 
t la 
moi n 
p rop 
a met 
ite 
a ren 
fond 
x au 
Le p 
61 es 
nta n 
dans 
/cm3 

c i ab 

ns d 



n verger 
de 45 a 

ce 
ement bas , 
e 

rappel 1 e 
pour 1 es 
phreatique 
es 

i re se 
carottes 
t la gamme 
able 

es ni veaux 
a i re dans 

ou 



espo 
s 1 e 
eat i 

va 1 
s bo 
Met 
res 
appa 
t e a 
e (p 
x 
ou re 

tex 
t pa 

1 es 
) ne 
1 e p 
u te 



et 



n d a i t 
s 

que 
eu r 
nne . 
es 
de 
rente 

ete 
our un 

enta ge 

turaux 
r 



ou r 
rrai n . 



i v 



TABLE OF CONTENTS 



Page 

INTRODUCTION - - 1 

MATERIALS AND METHODS - -- 1 

Field measurement of capillary rise 1 

GENERAL APPROACH - , 1 

CYLINDER DESIGN -- 3 

RAIN SHELTERS - 3 

SOIL MOISTURE SAMPLES - 3 

SOIL MOISTURE TENSION 6 

WATER TABLE DEPTH - 6 

Laboratory measurements 6 

SOIL SAMPLES - 6 

SATURATUED HYDRAULIC CONDUCTIVITY 7 

MOISTURE RETENTION CURVE 8 

Calculation of capillary rise from soil properties 8 

RESULTS AND DISCUSSION 12 

Field measurement of capillary rise 12 

Comparison with other sites 13 

Comparison of observed with calculated rise 23 

Correlations of hydraulic properties and derived 

parameters with other soil properties 24 

Influence of capillary rise on crop growth 30 

REFERENCES 32 

APPENDIX A 34 

LIST OF FIGURES 41 

LIST OF TABLES 42 

LIST OF COMMON SYMBOLS 44 



- 1 - 



INTRODUCTION 

In a recent study of factors affecting rooting depth of apple 
trees (Webster 1978), one orchard (D) was unusual in several 
respects. Although rooting was shallow there was no apparent 
barrier to deep rooting; tree performance was better than is 
associated with shallow rooting and, within the zone of root 
development, roots were exceptionally abundant. These features 
led to the speculation that capillary rise from a water table 
contributed significantly to available water in this orchard and 
to the more general speculation that the contribution of 
capillary rise to available water may be appreciable in some, 
years on similar sites in this region. 

Capillary rise is a dependable source of water for plant 
growth only under exceptional circumstances where natural or 
man-made features enable relatively stable water table depth 
regardless of weather conditions. In the more usual 
circumstances, where water table depth fluctuates widely 
depending upon weather conditions, capillary rise is of no 
practical interest. However, in the investigation of 
relationships between soil properties and crop performance, rise 
is of interest for research purposes if it can contribute to the 
water supply on some sites in some years because, if this 
contribution can not be estimated and segregated, it will be a 
source of experimental error. 

This study was undertaken (1) to examine the magnitude of 
capillary rise in this orchard (orchard D), (2) to characterize, 
to a limited extent, the hydraulic properties of the subsoil 
under this orchard and (3) to compare the hydraulic properties of 
samples from Orchard D with hydraulic properties of samples 
representing the range of soil texture commonly used for apple 
orchards in this region. 

MATERIALS AND METHODS 

The experimental field (Sheffield Farm, Kentville Research 
Station field B2; orchard block 23) was mapped as Somerset in the 
most recent soil survey (Cann et al. 1965), slopes northward with 
a gradient of approximately 1:50 and was planted to orchard in 
1962 at a spacing of 4.6 x 7.9 m. Capillary rise was measured at 
two vacant tree sites, row 2 tree 28 and row 2 tree 42. 

Field measurement of capillary rise 



GENERAL APPROACH At each of these two sites, two iron cylinders 
were installed, filled 5 cm above ground level with disturbed 
soil (to allow for settling), and covered with raised translucent 



- 2 - 



ran n 
remo 
fert 
cyl i 
soi 1 
subs 
i r r i 
cyl i 
Tabl 
of r 
tens 
cyl i 
to k 
appr 



she 
ved 
i 1 i z 
nder 

i n 
oil 
gati 
nder 
e 1) 
u n ; 
i on 
nder 
eep 
oxi m 



1 ters 
from 
er be 

(ope 
the s 
by me 
on. 
s , vo 

and 
Table 
(gaug 
s and 
soi 1 
ately 



Ear 
all cy 
i n g mi 
n ) res 
econd 
ans of 
After 
1 umet r 
measu r 

1). 
e ) was 

the c 
moi stu 

equal 



ly in 
1 i nde 
xed i 
ted d 
cy 1 i n 

a sh 
a cov 
ic so 
ed ag 
D u r i n 

moni 
1 osed 
re te 



each 
rs (t 
n the 
i rect 
der ( 
al 1 ow 
er of 
i 1 mo 
ai n a 
g the 
t ored 

cyl i 
n s i o n 



subs 
o p s o i 

top 
ly on 
cl ose 

i ron 

pi an 
i stur 
fter 

cou r 

at a 
nder 

i n t 



equent 
1 segre 
20 cm. 

the un 
d ) was 

t ray w 
t growt 
e was m 
a perio 
se of a 

depth 
was sub 
he cl os 



growi n 
gated ) 

The s 
di stur 
i sol at 
i th pr 
h was 
easu re 
d of 2 

run s 
of 40 
- i r r i g 
ed and 



g season soil was 

and replaced, 
oil in one 
bed subsoi 1 . The 
ed from the 
ovision for sub- 
established in the 
d (start to run; 
to 39 days (end 
oil moisture 
cm in both 
ated as necessary 
open cylinders 



Evapot ranspi rat i on from the closed cylinder was considered 
to be , 



EV r = W 



cs 



W 



ce 



+ I 



and from the open cylinder 



EVn = W 



os 



-W 



oe 



+ R 



(1) 



(2) 



where EV is evapot ranspi rat i on , the letters c, o, s and e refer 
to closed cylinder, open cylinder, start of run and end of run 
respectively, W is the soil moisture content (cm) within the 
cylinder, I is sub-irrigation and R is capillary rise. Assuming 
that EV C = EV , 



R = W cs 



W ce + 



I -(Wqs -Woe) 



(3). 



Class A pan evaporation records, for comparison with EV C , were 
from the Kentville weather station located about 7 km from the 
capillary rise sites. 

It was not practical to generate soil moisture tension in 
the cylinders with apple trees and a low-growing stand of closely 
spaced plants was therefore used for this purpose. For example, 
moisture extraction by a tree advances outward and downward from 
the trunk in a somewhat irregular way depending upon root 
location as opposed to the more uniform moisture extraction by a 
stand of closely spaced plants. Uniform extraction was needed in 
order to monitor moisture tension and maintain equal availability 
of water in the closed and open cylinders. Further, the 
assumption of equal water usage from the closed and open 
cylinders is more likely to be approximated by a closely spaced 
stand of many plants than by one tree or several trees per 
cylinder. However, the results obtained by using a ground cover 
should apply equally to trees in the sense that flow, induced by 
a given moisture tension gradient, will be indifferent to the 
cause of the gradient. 



- 3 - 



CYLINDER DESIGN Cylinders were 55 cm long, 107 cm in diameter, 
coated with asphalt emulsion paint and installed June 1977, 30 cm 
apart in the tree row to a depth of 50 cm by digging a hole, 
setting the cylinders in place and refilling. The tray for the 
closed cylinder, also asphalt coated, was 114 cm in diameter with 
a 10-cm high rim and a central drain connected by pipe to a shut- 
off, gravel -fi 1 1 ed 30 L drain pit and a drain observation port 
(Fig. 1). The shut-off was closed during a run and left open 
over the dormant season so that free water, following periods of 
high water table, could drain from the tray. 

The bottom of the tray was covered with 3 cm of medium sand 
and the cylinder rested on blocks 3 cm above the tray. The space 
between the cylinder wall and the tray rim was filled with pea 
gravel and the rim to cylinder gap was sheathed with a strip of 
0.7 mm saran screen. A vertical 3.8 cm (ID) iron pipe, secured 
to the cylinder wall and leading to the bottom of the tray, was 
used for addition of sub-irrigation water, usually in amounts of 
2 or 4 L. Water was added rapidly so as to get maximum flow 
around the rim and across the tray bottom. A second vertical 
pipe of 1.3 cm diameter copper was used with a bubbler tube to 
monitor free water height near the tray rim during addition of 
sub-irrigation water to insure that water did not overflow the 
tray rim. 



i n r 

soil 

soi 1 

wate 

was 

tube 

pipe 

abov 

di am 

on t 

spac 

of t 

oute 



Thi 
i m p 

nea 

on 
r di 
capp 

i n 
. T 
e th 
eter 
he s 
ed 4 
he o 
r ri 



s me 
1 us 
rest 
the 
st ri 
ed a 
ring 
hese 
e t r 
) a n 
i de 
cm 
uter 
ng. 



thod 
sand 

the 
oppos 
b u t i o 
nd tw 
s of 

ri ng 
ay bo 
d fur 
di st a 
apart 

ring 



of sub- 
at bott 
s u b - i rr 
i t e s i d 
n t he b 

conce 
45 and 

s were 
ttom, p 
n i shed 

1 to th 
on the 
and 2 



irrig 
om) 1 
i g a t i 
e of 
ottom 
n t ri c 
85 cm 
suppo 
erf or 
with 
e dow 
i nne 
cm ap 



at i on t 
ed to u 
on down 
the cyl 
end of 
soft c 
d i a m e t 
rted 1 e 
ated on 
upri ght 
n pipe 
r ring 
a rt on 



hat was 
nequal w 

pi pe be 
i n d e r . 

the sub 
opper tu 
er ) we re 
vel , wi t 

the 1 ow 

and loo 
(Fig. 2) 
and the 
the d i s t 



used 
ater 
i ng m 
To pr 
- i r r i 
bes ( 
atta 
h the 
er si 
sely 
. Pe 
proxi 
al 2/ 



in 1 97 
d i s t r i 
ore mo 
o v i d e 
gat i on 
1.3cm 
ched t 
1 ower 
de (1 . 
capped 
rf orat 
mal 1/ 
3 port 



7 (g 

buti 
i st 
uni f 

dow 

di a 
o th 

si d 
6 mm 

ai r 
ions 
3 po 
i on 



ravel 
on ; 
than 
orm 

n pipe 
meter 
e down 
e 2 cm 

vent s 
were 
r t i o n 
of the 



RAIN SHELTERS Rain shelters were constructed from two panels 
(133 x 366 cm) of translucent corrugated fiberglass greenhouse 
covering secured to a light wooden frame to form a pitched 
shelter 368 cm long, 256 cm wide and with a pitch of 15 cm from 
each side to the center ridge and mounted with the ridge 90 and 
75 cm above ground level on the up and down slope ends 
respectively. Rain, collected from gutters on the down slope 
end, was conveyed by pipe a further 3 m down slope. 



SOIL MOISTURE SAMPLES A series of gouge shaped sampling augers 



- 4 - 



SCALE 



50cm 



lf n n 




Fig. 1. Design of closed cylinder. Tray (A) with gravel-filled 
rim, and 3 cm of medium sand provided with central drain, shutoff 
(B), gravel-filled drain pit (C) and drain observations port (D) 
and sheathed with a strip of 0.7 mm saran screen (E). Pipes F 
and G were for addition of sub- irrigation water and measurement 
of water height near the tray rim during addition respectively. 



SCALE 



50cm 




Fig. 2. Top view of perforated con- 
centric copper water distribution tubes 
showing vertical iron down pipe (F) and 
loosely capped air vents (H). 



- 5 



Cn 


+-> 


r-» 


>> CD CL 


cn 


U CJ CD O 


i— i 


SI <=C CO CVJ 


CVJ 


LO 1— • O 


=**= 


CVJ CVJ t-H 



CO 

O) 
4-> 



O 

2 
4-> 

+-> 

13 



+-> 

C 
•f— 

E 

£_ 
O) 

+-> 

<V 

-o 

a; 
00 



<0 



CD- 
is 
o 



o 

to 

t_ 

13 

cj 
o 

•(-> 

t- 

13 
Q. 

■a 
fl3 
<V 

Z5 

■a 
cu 

.c 
o 

CO 



qj 






Cn 


>3 


r-~. 


>>r— cn 


Cn 


<o 3 3 w 


-— 


S0<(0 


A 


LO O 1 


CVJ 


CVJ CVJ CVJ 


=«= 





00 ^r cn 00 co 
cvj «3- , — to , — 

• • • • • 

00 CVJ l-» CO o 



cx> lo cvj co cvj 
• • • • • 

r-'tCOCOO 



cn 


>3-l-> 


r- 


>,.— OL 


cn 


fl 3 (HOI 


■— 


Z onco 


n 


Lnrv^t 


i — 1 


CVJ CVJ 



COMN'vl-O 
CO CO CVJ C7> CVJ 



cri lo i^«. r-^ o 



=tfc 



cn 


>> >i 


i*-» 


>>r- .— 


Cn 


13 C3 ZJ CO 


■— 


21 "-3 ■"D CVJ 


«\ 


lo ^d- r*- 


■ — 1 


CVJ CVJ 



o lo co r-- cvj 



IT) CTl N O 



=*fc 



r-« 


>•> 4-> 


1*^ 


1 — D)H 


0-1 


CS C3 CU ^T 


"— 


O e£ CO CVJ 


n 


lo cn cvj 


CVJ 


cvj 


=*fc 





r-. cn co co co 

O LO O LO CVJ 



H ^-OifiO 



r~~- 


>> +-> 


r»» 


. — CD CL 


en 


cs =s aj cn 


•— 


-3 <: 00 cvj 


n 


LO ^f CVJ 


i — 1 


CVJ 



, — r^ co cn cvj 



,— <^f- 00 LO O 



en co 
LO lo 



lo o 



CO LO 
LO LO 



CVJ o 



CO CO 
LO LO 



CO o 



CO LO 
CO LO 



CO LO 

cn lo 



en o 



CVJ LO 



CVJ C3 














CVJ LO 


• 


co 




£_ 


c-l CVJ 


1 — 


, — 


LO 


0) 


• • 


+1 


• 


l~-- 


•O 


00 r*» 


r-- 

• 

CVJ 

O 







5/cy 1 i n 
s . 


O CVJ 


• 


LT> 




<* +_> 


CO t-H 


1 — 


1 — 


LO 


c 


• • 


+1 


• 


LO 


"O U 


1— CO 


1^- 

• 

10 


O 




s pi ante 
to 25 pi 


00 


• 


co 




cn 


LO LO 


r— 


1 — 


CVJ 


c: -0 


• • 


+ 1 


• 


LO 


•r- cu 


CO LO 


O 

• 

LO 

LO 


O 




■ — tz 
■0 c 

CU -r- 

to +-> 

•r— •> 
r— 0J 


CO CO 


• 


LO 




O C 


LO LO 


1 — 


, 


LO 


O C3 


• • 


+ 1 


• 


**- 


O >~3 


CVJ CO 


CO 

• 
CO 


O 




O 

C_ , — 
CO r— 



00 r^ 

CO r— 



r— O 



LO 
• CO 

1— 1— en 

+1 .1^ 
^- o 



cvj r- 

«3- t— 



o r*. 



• LO 

1 — 1 — cvj 

+ 1 * LO 
CO o 



-a 
a> 

4-> 

c: 

. — r~~- 
o. en 
to 1 — 
c: 

as cu 
J- c: 

+-> Z3 

-3 
■fr 
-a lo 

O r-H 
CO 

























E 


















-a 




















a 


















a> 




















* — 


















1 — 






































1 — 




















c 












*-^i 






13 










»—>» 

















' — • 




>1 


CU 






-t-> 










E 








' — » 


•1— 






E 




13 


to 






LO 










u 




• ■ 




>> 


-t-> 






a — 




"O 


C5 






c 










— 


1 — «■ 


E 




03 


13 






"— - E 


- — ^ 


"»«, 








•r— 












E 


O 




-a 


t_ 


c: 







E 


E 


t_ 
















4-> 


O 


» — 


- — - 


\ 


O 


13 




+-> — 


O 


<_) 


0) 






CO 










L 







E 


E 


Q. 


CL 




t_ 







4-> 






L. 


c 






L_ 


T3 




-a 


O 


u 


13 






13 "0 






13 






QJ 


3 






CU 


+J 


~a 


c 


■> — 


— 


> 


<c 




+-> C 


CU 




2 




t_ 


+-> 


L_ 


c 


to 


-a 


00 


QJ 


0J 






CU 




£_ 


to a> 


to 








13 


1 — 




Z3 


>> 


c 




-0 




QJ 






to 


0J 




•1— 




<4- 


£_ 


cu 


QJ 


4- 


L_ 


13 


■1— 


+J 


"O 


+J 


oo 




c 


to 


"O 


4-> +-> 


1_ 




O 


aj 


>t 


to 





l+_ 


XJ 


>> 


T3 


10 


13 


Z3 




13 
CL 


13 


c 


U 13 


>•> 




&^- 


> 



" 




+-> 


O 


4- 


u 


C_ 


c_ 


(_ 


L- 






O 


. — 


t_ t_ 


t_ 




1 


u 


aj 


c 


L_ 




O 




QJ 


aj 


QJ 


0J 




< 


\ 


>> 


aj aj 


13 








-(-> 


•(— 


T3 


T3 




-a 


-t-> 


+j 


4-J 


4-9 






QJ 


u 


+-> +-> 


r— 






T3 


•1— 


fa 


4-> 


cr 


O 


aj 


ra 


<T3 


n3 


13 




to 


to 




13 13 


1 — 






C 


co 


cc 


t/J 


UJ 


^ 


to 





3 


ZS. 


3 


3 




to 

13 




=3 


cu 

Q. 
O 


3 3 


•r— 
Q. 
13 

<_> 






3 
O 

CO 



to 
+J 





to 


aj 


•1— 


a 


to 


c 


c 


QJ 


CD 


"O 


4-> 


•«— 


13 


H- 


t_ 


c: 


o_ 



O 




O 


13 




O 


S^ 


Q_ 


LO 




cn 


>1 




1— 




U- 


QJ 


OJ 


00 


•r— 


•j— 


-C 


cc: 


CJ 


•• -H- 



- 6 - 



(Eijkelkamp B.V.) 40, 35, 30, 25 and 20 mm diameter were used to 
take four soil samples, in relatively undisturbed condition, from 
depths of 0-10, 10-20, 20-30, 30-40, and 40-50 cm respectively, 
samples from opposite quarter positions being pooled. After 
trimming to know volume the samples were oven dried at 105°C for 
calculation of volumetric soil moisture and cm water over the 50 
cm prof i 1 e. 



as 



The 95% confidence limits for capillary rise were calculated 



95% limits = 2.776 [z(rep 1 -rep 2) 2 /4] ' 5 



with the duplicate determinations (rep 1 and rep 2) of W CS ,W 



W 



OS 



and W 



oe 



entered into the summation. 



ce» 



SOIL MOISTURE TENSION Tensiometer gauges were calibrated at four 
points (10, 20, 40 and 60 centibars) and two tensiometers were 
installed in each cylinder at a depth of 40 cm. Tensiometer 
readings, taken three to six times per week depending upon 
anticipted rate of change or need for irrigation were corrected 
for gauge bias and height from gauge to ceramic element. 

WATER TABLE DEPTH Five observation wells were installed July, 
1977 along a line tangential to and about 1 m from each open 
cylinder by drilling holes 4.7 cm in diameter and about 0.5 m 
apart to depths of 60, 100, 170, 200 and 250 cm. These wells 
were cased with ABS pipe (3.8 cm ID, 4.76 cm 0D) that had been 
slotted over depths of 30-54, 70-94, 100-164, 170-194 and 150-244 
cm respectively with staggered 3-mm wide saw cuts at quarter 
positions. Wells were covered with a loose fitting cap. 

Depth readings were taken with a bubbler rod three to six 
times per week, corrected with reference to the average soil 
surface elevation of the five wells at each site and averaged for 
a given day after editing out readings indicative of sluggish 
response (the bottoms of some wells became impermeable and 
outflow from these wells was very slow after the water table had 
dropped below the slotted section). 

Laboratory measurements 

SOIL SAMPLES Cores of relatively undisturbed soil, 7.6 cm in 
diameter and 3 cm long, were taken within 1 m of the open 
cylinder at the two capillary rise sites and at eight other 
locations (Table 2), using a sampler similar to the one described 
by Swanson (1950). These eight locations were selected so that 
the range of soil texture common in apple orchard soils of the 
region would be represented. Both ends of the core samples were 
trimmed flush with the lucite retaining cylinders and the lower 
end was covered with 53 ym nylon mesh (B. and S. H. Thompson 



- 7 - 



Table 2. Source of core samples 



Sample no. 



Depth No. of 
(cm top of core) cores 



Texture of 
Soi 1 type + sampl e 



9 
11 

8 

7 
55 
99 

5 
10 
13 
12 



(Site 
(Site 



#1) 
#2) 



45 
65 
65 
85 

70,80 
85 

65,80 
85 
85 
45 



N i c t a u x 
Berwi ck 
Somerset 
Debert 
Somerset 
Berwi ck 
Somerset 
Berwi ck 
Kent vi 1 1 e 
Mi ddl eton 



sand 

sand 

sand 

1 oamy 

sandy 

sandy 

sandy 

sandy 

sandy 



sand 
loam 
1 oam 
1 oam 
1 oam 
loam 



1 oam-cl ay 
1 oam 



.Soil series as described in Cann et al . (1965). 
^Sites at which capillary rise was measured in field. 



and Co. Ltd., 235 Montpellier Blvd., Montreal) held in place with 
an elastic band. This nylon mesh was not removed until 
conductivity and moisture retention determinations were 
completed. Additional soil was collected at the same time for 
determination of moisture retention at 15 bar, particle density 
(pycnometer, method 2.25 in McKeague ed. 1978), percent carbon 
(modified Wal kl ey-Bl ack , method 3.613 in McKeague ed. 1978) and 
particle size distribution (pipet method, method 2.111 in 
McKeague ed. 1 978). 

SATURATED HYDRAULIC CONDUCTIVITY An empty core retaining 
cylinder 7.6 cm long and resting on a 1 -mm wire screen disc (8.2 
cm diam.) was fastened to the upper end of the core sample with a 
wide elastic band. Soil cores were then placed on a rigid 
support consisting of a disc of 0.5-mm nylon mesh overlying a 
perforated lucite disc provided with wire handles so that the 
entire assembly could be moved without disturbing the samples. 
Samples were moistened from below with deaerated 0.01 N CaS04 
containing 1 mL of 40% formaldehyde per litre and soaked in this 
solution a further 24 hours. After transfer of the core assembly 
to a funnel, hydraulic conductivity was determined at constant 
head as described by Klute (1965; pp 214-215) using a Mariotte 
bottle for head control. Average hydraulic conductivity was 
calculated as a geometric mean and the associated standard 
deviation accordingly implies multiplication and division not 
addition and subtraction. 



- 8 



MOIST 

deter 

cal cu 

tensi 

deter 

water 

(McKe 

times 

7.6 c 

wi 1 1 

(Rich 

brack 

[3], 

[30], 

folio 

was a 

and w 

curve 

mean 



URE RET 
m i n a t i o 
1 at i on 
on tabl 
m i n a t i o 

as des 
ague ed 

approx 
m long; 
vary ap 
ards 19 
eted mi 
10 [4], 

225 [3 
wed Top 

2-cm 1 
as cove 
s over 
at each 



ENTION C 
n of sat 
of moist 
e (al ong 
n of moi 
c r i b e d i 
. 1978; 
i m a t e 1 y 

the rat 
proximat 
65; p 13 
nimum eq 

20 [4], 
8], 300 
p and Ze 
ayer of 
red with 
two or m 

tensi on 



URVE 
urate 
u re r 
wi th 
stu re 
n Can 
PP 43 
1/6 a 
i o n a 1 
ely w 

6). 
u i 1 i b 

40 [ 
[45] 
bchuk 
9.5 u 

53 u 
ore c 

poi n 



Samp 
d hyd 
etent 

cone 

rete 
adi an 
-44) 
s 1 on 
e bei 
ith t 
Tensi 
rati o 
10], 
and 5 

(197 
m al u 
m ny 1 
ores 
t. 



1 es we 
r a u 1 i c 
ion at 
u rrent 
n t i o n 
Soc. 
but w i 
g as t 
ng tha 
he squ 
o n s in 
n time 
60 [10 
00 [52 
9) exc 
m i n u m 
on mes 
were c 



re wei 
condu 
zero 
ly soa 
over t 
Soil S 
t h m i n 
hose r 
t ti me 
are of 
cm wa 
s i n h 
], 80 
]. Th 
ept th 
oxi de 
h. Me 
al cul a 



ghed imm 
cti vi ty , 
tensi on , 
ked bl an 
he range 
ci . Meth 
i m u m e q u 
ecommend 
to reac 
sample 
ter foil 
ours wer 
[15], 10 
e tensi o 
at the t 
powder f 
an moi st 
ted as t 



ediately after 

for 

and moved to a 
k cores) for 

of 5-500 cm 
ods Manual 
i 1 i b r a t i o n 
ed for cores 
h equi 1 i bri urn 
heights 
owed by 

e as f ol 1 ows : 5 
[24], 150 
n table de-sign 
en si on medium 
or all tensi ons 
ure retention 
he arithmetic 



Retention at 1 bar (14.5 psi) was measured using a ceramic 
pressure plate with a thin layer of 9.5 ym aluminum oxide powder 
between the core sample and the ceramic surface. Retention at 15 
bar was determined with the fine fraction (< 2-mm) of an air 
dried sub-sample using a ceramic pressure plate and expressed on 
a fractional volume basis after adjustment for coarse fraction 
(> 2-mm) content and core bulk density, i.e. 

015 bar = w 15 bar x f x &b 

where 0^5 b ar is adjusted water content (cm3/cm3) at 15 bar, 
w 1 5 bar 1S 9 water retained at 15 bar per g fine soil, f is q 
fine soil per g unsieved soil, D^ is core bulk density (g/cm3), 
fine soil refers to the fraction passing a 2-mm sieve and all 
soil weights are on an oven-dry (105°C) basis. It is assumed 
that the coarse fraction will retain a negligible amount of water 
(Richards 1965; p 136) and further assumed that water retained 
per g fine soil at 15 bar will not be influenced by degree of 
compact i on . 

Calculation of capillary rise from soil properties 

The procedure for calculating the capillary rise of water 
from the water table was based on Gardner's (1958) steady-state 
solution to the soil water flow equation. The calculations were 
carried out on a daily basis. These calculations required a 
relationship between hydraulic conductivity, K (cm/min), and 
tension, h (cm water). Measurements of K were made only at 
saturation and it was necessary to use a predictive method for 
estimates of K at lower water contents and higher tensions. A 
procedure developed by Brooks and Corey (1964) was used for this 



- 9 - 



purpose. (Orientation studies had indicated that an alternative 
procedure, the Millington and Quirk model (Bouwer and Jackson 
1974), tended to overestimate the decrease in K with increase in 
tension.) The Brooks and Corey procedure was modified slightly, 
as described below (Eqs. 4 to 8) to allow for gas filled voids at 
zero tension and to accommodate the gradual decrease in water 
content (0) of these undisturbed cores as tension was increased 
from zero to the air entry value (hb). 

Effective saturation (E s ) was calculated as 

Es = ((e/o s ) - S P )/(1 - S r ) (4) 

where Q s is the water content at zero tension and S r is residual 
saturation. The value of S r was selected so that the absolute 
value of the correlation coefficient between In E s and In h was 
maximal after exclusion of the two to four points at low suction 
(0 and 5 cm or 0-20 cm water). The relationship In E s vs In h 
was always non-linear at low tension. 

After selecting S r as above the moisture retention curve 
(MRC) can be represented by 



In E s = Xln hb - Xln h for h >hb 



(5) 



where h is tension in cm water, hb is the air entry value and ^ 
is a pore size distribution index. Larger values of X indicate 
narrower distributions. In equivalent form Eq. 5 becomes 



Es = (hb/h) 



X 



(6) 



Brooks and Corey (1964) found that the hydraulic 
conductivity (K) was represented well by 



K = K c E, (2 + 3X)/X 



(7) 



where K s is the saturated hydraulic conductivity. The 
substitution of Eq. 6 into Eq. 7 for E s gives 

K = K s (h b /h) (2 + 3X) for h >hb ( 8) 

Gardner (1958) used a three parameter equation of the form 

K = a/ (h" + b) (9). 

Least squares estimates of the constants a, n and b of Eq. 9 were 
obtained by fitting K, calculated by entering Eq. 7 with E s from 
Eq. 4 at the 12 h values of the MRC, into the logarithmic 
transform of Eq. 9, i.e. 



lnK=lna-ln(h n +b) 



(10). 



- 10 - 



At saturation, h = was replaced by an arbitrarily 



of h (usually 0.001 
separately i.e. a/b 



cm water ) . 
was not set 



The constants a and 
equal to K s . 



smal 1 value 
b were fitted 



Digression 

The standard Brooks 

with E s from Eq. 6) 

value with K = 

of Eq. 9 is zero, i 



and Corey model (Eq. 8 or Eq. 7 entered 
assumes a sharply defined air entry 
K s at h <hb and assumes that the constant b 
e. Eq. 9 is equivalent to Eq. 8 if 



a = K s h b (2 + 3X) 

n = 2 + 3A 



b = 0. 
In contrast, entering Eq. 7 with E s from Eq. 4 allows K s to 
decrease gradually as tension is increased from zero to hb, 
consistent with the observed decrease in between h = and 
h = hb« This approach yielded estimates of K at hb that 
were intermediate between those obtained by Eq. 8 and by the 
Mi 1 1 ington and Quirk model (Bouwer and Jackson 1974). 



Equation 9 was then used to calculate K at 
values of h to enable numerical integration of 

r n max 
Z = J dh/(l+q/K) 



closely spaced 



(11) 



(Gardner 1958) where Z is the height above a water table at which 
a tension of h max would maintain a stead upward flux of q. 
Tension increments, for the j steps in this integration over the 
range h = to h = h max , were increased exponentially such that 
given step i , 



for a 



h-j = h-j _i + P" 1 and 



h 



max 



= h 



j-l 



+ PJ 



(12) 
(13), 



j being the maximum value of i and P being some value that 

satisfied Eq. 13 for the given values of j and h max . For 

integration of Eq. 11 the height above the water table (Z-j) at 
which the tension was h-j was calculated as 

Zi = Z-j.! + (hi - hi_i) x (K i _ 1 /(q + Ki_-|) + Ki/(q + K i ) } / 2 (14) 

Estimates of q for the experimental sites based on measured 
soil properties, water table depths and daily soil water tensions 
were the desired result to allow comparisons with the measured 
capillary rise for the several experimental periods. Equation 
11, however, can not be solved explicitly for q. It was 
necessary to integrate for a number of q values and iterate to a 



- 11 - 



value of q that satisfied h max (soil moisture tension at 40 cm 
depth in the open cylinder) and Z (depth to water table minus 40 
cm). The integration was calculated on a daily basis with j = 
50. Daily values of h max and Z were obtained by linear 
interpolation between observation dates (two to six times per 
week). The predicted or calculated rise was the sum of the daily 
q estimates over the experimental period. 

Detailed calculations of Z:q relationships for the 10 sites 
were carried out for q values of biological interest, in order to 
compare the two measured sites with the remaining eight sites on 
which soil properties were determined. In these calculations 
using Eq. 11, h max was set at 500 or 4000 cm water, and q was 
varied from 0.01 to 1.0 cm/day in 10 steps where qj+] = q-j l.O^* 2 . 
The integrations were carried out in 500 steps (i.e. j = 500).. The 
resultant matrix of numeric values (10 sites x 11 flux rates x 2 

h, 

each h max according to 



lmax ) was further simplified by fitting q vs Z for each site and 



In q (cm/day) = a' + n'ln Z (cm) 

The values of a' and n' were calculated by 
estimates. For the conditions where 1 + qb/a - 
given from Eq. 9) with a and q expressed in the 
is nearly linear with In Z and n' - -n of Eq. 9 
Eqs. 18 to 21). Note that the constants a' and 
beyond the range of q under consideration (0.01 
The curvature of the In q vs In Z relationship 



increases and 
relative to h 
as q is decreased ) . 
rel at i onsh i p occurs 



also as q is 
max ( z cannot 



decreased such that Z 
exceed h max and it wil 



Thus, non-linearity of In 
at extremes and beyond the 



(15). 

1 east squares 
1 , (a and b are 
same units, In q 
(Gardner 1958; 
n ' do not apply 
to 1.0 cm/day ) . 

i ncreases as q 
is large 

approach h max 

q vs 1 n Z 

range of q used. 



The constants a 1 and n 1 have the advantage that they 
characterize capillary rise properties in a form analogous to 
Gardner's (1958) Eqs. 18 to 21 and express q as a function of Z. 
For the purposes of correlation with other soil properties such 
as percent sand, the coefficients a' and n 1 have the disadvantage 
that they were highly correlated (r = -0.98) with each other. In 
regression equations of the form Y = C] + C£X, the degree of 
correlation between C] and C£ will be minimal if X is replaced by 
X-X (Draper and Smith 1967, p 22). By reversing the variables in 
Eq. 15 and cha ngin g the scale in q to mm/day, mean In q = and 
In q = In q - In q. Thus the relationship is written 



In Z (cm) = a" + n"ln q (mm/day) 



(16), 



where the constants a" and n" are determined by least squares 
estimates and the degree of correlation between a" and n" was 
reduced ( + 0.67). 



- 12 



Equation 16 has the disadvantage that it is written with 
In Z as the dependent variable when in practice In q is the 
dependent variable (i.e. one is interested in the effect of Z on 
q and not vice versa). The correlation between In q and In Z 
over the range 0.1 to 10 mm/day is almost unity so the reversal 
of variables offers no problem mathematically but does imply the 
awkwardness of thinking backwards. The constant, a" is In Z when 
q = 1 mm/day and can be related to other soil properties. This 
is in contrast with a' which could not be considered separately 
from n ' . 



RESULTS AND DISCUSSION 

Field measurement of capillary rise 

Mean capillary rise over the experimental periods in 1977 and 
1979 ranged from 0.13 to 0.18 cm/day (Table 1, Fig. 3) and 
supplied from 45 to 79% of the water used during these periods. 
In view of the relatively large 95% confidence limits surrounding 
capillary rise measurements (Table 1) these figures (cm/day, 
percent of water use) are of course approximate but show that 
rise was a significant source of water during these periods. 
Water usage during these periods was 53 to 66% of Class A pan 
evaporation, slightly lower than would be expected under normal 
field conditions (80%, assuming a crop coefficient of 1 and a pan 
coefficient of 0.8; Doorenbos and Kassam 1979, Tables 17 and 18), 
perhaps because all water was supplied from below and evaporation 
from the soil surface would be less than normal. 



The tensiometer record is also indicative of substantial 
capillary rise (Fig. 3). Soil moisture tension in the open 
cylinders fell rapidly (or increased less rapidly; site 1 1977) 
following an abrupt decrease in water table depth. In 1979 the 
cylinders were rain-free for 14 weeks and supported a closely 
planted vigorous annual crop but soil moisture tension at a depth 
of 40 cm in the open cylinders exceeded 50 cb only briefly during 
this period (Fig. 3). 

Water table depths were unusually low in 1978 (rainfall for 
the month of May was 32% of normal) and no measurement of rise 
was possible. Soil moisture tensions rapidly increased beyond 
the functional range of tensiometers (Fig. 3) and plant growth in 
the cylinders was very poor. This experience in 1978 is a 
reminder that capillary rise is a reliable source of water for 
plant growth only if water table depth is reliably stable within 
certain limits. 



- 13 



SITE 



SITE 2 



z 
o 
(/) 

z 

UJ 



llJ 



00 

< 



UJ 

(- 

5 



50 



100 - 



£ 200 



z 
g 

V) 

z 

UJ 




50 



E 

UJ 

_i 

03 100 

< 



200 



1979 



v^s*s 



QI5 






013 



v V 



JUNE JULY 



AUG. SEPT. 




Fig. 3. Soil moisture tension in centibars at a depth of 40 cm 
(open cylinders) and depth to water table at sites 1 and 2 over 
portions of three growing seasons. Numbers within the figure are 
mean capillary rise (cm/day) measured over the period indicated 
by the underlying line. 



Comparison with other sites 

Having established that capillary rise can be substantial in the 
experimental orchard in some years it is pertinent to ask if the 
hydraulic properties of the subsoil in this orchard are unusually 
conducive to capillary rise. Tables 3 to 8 relate, directly or 
indirectly to this question and show that the hydraulic 
properties of the capillary rise sites (sample nos. 55 and 5) are 
not unusual but fall well within the range of core samples from 
the other ei ght sites. 

Saturated hydraulic conductivity of samples from the 10 
sites ranged from 1.94 to 0.0037 cm/min (Table 3) and was, for 
most sites, similar to K s that would be expected on the basis of 



- 14 - 



Table 3. Moisture retention and other properties 
from 10 sites 



of core samples 



Li ne 

no. Sampl e no. 

1 Bui k density ± SD (g/cm 3 ) 

2 Particle density (g/cm 3 ) 

3 Total porosity (%) 

4 Fine (% <2 mm) 

5 Carbon (%) 

6 Sand (% 2-0.05 mm) 

7 2-1 mm (%) 

8 1-0.5 mm (X) 

9 0.5-0.25 mm (%) 

10 0.2 5-0.1 mm (%) 

11 0.1 -0.05 mm (%) 

12 Silt (% 0.05-0.00 2 mm) 

13 Clay (% <0.002 mm) 



14 
15 
16 
1 7 
18 
19 
20 
21 
22 
23 
24 
25 
26 
27 
Hydra 

(K s ; 

K s _pr 



water ) 



"^Geom 
mu 1 1 

§Pred 
base 
bei n 
tens 
tens 



Tension 
( h ; cm of 

5 
10 
20 
40 
60 
80 
100 
1 50 
225 
300 
500 
1020 (1 bar) 
1 5300 (1 5 bar) 
ulic conductivity! 
cm/mi n, saturated) 
e d i c t e d ^ 

a n t i 1 o g 



SD' 



1 .56 
2.673 
41 .6 

65.1 
0.1 5 

93.1 

25.4 

26.8 

17.8 

18.7 
4.4 

1 .6 
5.4 



11 

1 .61 
2.695 
40.3 

95.3 
0.22 

92.6 

11.0 

19.4 

23.4 

30.4 
8.4 
2.1 
5.3 



8 

1 .67±0.04 
2.710 
38.4 

99.9 

0.1 3 
91 .5 

5.0 
12.9 
25.2 
29.9 
18.6 
1 .7 

6.8 



1 .67±0.09 
2.743 
39.1 

99.9 

0.02 
84.5 . 

0.5 

4.4 

10.5 

50.4 

18.7 

7.4 

8.1 



Moisture retention (0; cm 3 /cm 3 ) 



0.367 
0.329 
0.308 
0.295 
0.1 62 
0.110 
0.089 
0.076 
0.064 
0.055 
048 
044 
038 
01 9 



1 



94 
74 



0.348 
0.345 
0.338 
0.300 
0.216 
0.1 75 
0.1 51 
0.1 33 
0.1 15 
0.1 05 
0.098 
0.094 
0.086 
0.027 

0.45 
1.12 



0.330 
0.308 
0.297 
0.283 
0.235 
0.183 
0.162 
0.1 40 
0.108 
0.088 
0.082 
0.072 
0.066 
0.043 

0.1 6±1 .4 
0.55 



0.359 
0.342 
0.326 
0.311 
0.270 
0.232 
0.220 
0.206 
0.182 
0.1 55 
0.141 
0.1 24 
0.11 
0.071 

0.11+2.1 
0.18 



etnc 
i p 1 i c 
i c t e d 
d on 
g, In 
i on , 
i on ; 



mean : 
a t i o n 
K. 



and d i v i 

was ca 1 cul 

data of Mason 

K ( c m / m i n ) = 

over the range 

sites 8, 9 and 



SD; the ± 
si on . 

ated from a 

et al . (1 95 

-5. 91 +0.2 

of 2.4 to 

11 fell ou 



sign fn this ca se i m p 1 i e s 

linear regression equation 
7, Table 3) this equation 
64% air porosity at 60 cm 
18.4% air porosity at 60 cm 
t si de of this range. 



- 15 - 



Table 3. (continued) 



Line 
no. 



55 



99 



10 



13 



12 



1 1.75±0.04 1.99+0.03 1.83±0.05 1.67±0.07 1.88±0.03 1.78±0.07 

2 2.703 2.709 2.684 2.742 2.699 2.759 

3 35.3 26.5 31.8 39.1 30.3 35.5 



4 

5 

6 

7 

8 

9 

10 

11 

12 

13 



96.2 

0.02 
73.7 
13.4 
20.9 
15.2 
16.7 
7.6 
17.1 
9.1 



88.5 
0.01 
67.3 
8.4 
14.8 
12.2 
19.0 
13.1 
19.6 
13.1 



96.2 

0.04 
66.4 
12.0 
1 7.6 
13.1 
16.1 
7.5 
1 5.1 
18.6 



98.7 

0.01 

55.4 

1.3 

3.1 

5.5 

27.0 

18.5 

29.2 

15.4 



Moisture retention (0; cm^/cm^) 



92.4 
0.00 

54.6 
8.6 

11 .7 
9.0 

13.8 

11.5 

29.1 

16.3 



97.1 

0.06 

44.5 

5.7 

7.3 

6.3 

11.7 

13.5 

28.2 

27.4 



14 

15 



16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
26 
27 



0.320 
0.317 
0.302 
0.269 
0.232 
0.208 
0.202 
0.198 
0.1 92 
0.184 
0.1 79 
0.168 
0.1 54 
0.066 

0.11±1 
0.1 2 



0.284 
0.269 
0.260 
0.250 
0.231 
0.223 
0.21 7 
0.21 2 
0.205 
1 99 
1 95 
187 
1 76 
067 



0.324 
0.323 
0.31 7 
0.299 
0.272 
0.256 
0.249 
0.243 
0.235 
0.226 
0.220 
0.209 
0.1 97 
0.1 01 



0.0051 ±1.2 0.025±1 
0.0082 0.014 



0.357 
0.353 
0.345 
0.334 
0.31 2 
0.293 
0.277 
0.264 
0.247 
0.231 
0.21 9 
0.205 
0.188 
0.093 

0.020±2 
0.036 



0.299 

0.289 

0.281 

0.276 

0.258 

0.250 

0.244 

0.238 

0.232 

0.21 5 

0.221 

0.213 

0.1 99 

0.1 01 

0.0037±1 
0.01 1 



0.349 
0.338 
0.337 
0.328 
0.319 
0.314 
0.310 
0.306 
0.300 
0.293 
0.289 
0.284 
0.272 
0.145 

! 0.0038+2.4 
0.0080 



Retention at 225 cm (sample 13) was 
excluded from further analysis. 



considered to be in error and 



- 16 - 



air porosity at 60 cm tension (Mason et al . 1957; Table 3). 
There was an approximately linear relationship between percent 
sand and In K s ; with samples arranged in decreasing order of 
percent sand they were, excepting sample 99, also arranged 
approximately in order of decreasing K s . Sample 99 had an 
unusually high bulk density. 

Absolute values of the correlation coefficients between In 
E s and In h, over the range h = 10, 20 or 40 to h = 500, 
approached unity (Table 4) indicating that A and h D (Eq. 5) 
adequately described the moisture retention curves of these 
samples. In most cases the absolute value of this correlation 
coefficient increased as S r was decreased, reached a maximum 
value and then decreased (after excluding two to four data points 
at the low tension end of the moisture retention curve), this • 
value of S r at which a maximum was reached being the selected 
value of S r . 



D i g r e s s i 



Howe 
reac 
the 
to r 

Sr = 
more 

poi n 

S r f 

vari 

but 

with 

was 

bei n 

with 

val u 

7. 

12, 

for 

p ro v 

resp 

1 2 w 

va 1 u 

cons 

base 

resp 



ver 
hed. 

com 
each 

0.1 

or 
ts o 
or s 
ous 
capi 
i n t 
s i m i 
g an 

Sr 
e of 

The 

fal 1 

samp 

i ded 

ect i 

ere 

es a 

tant 

d on 

ect i 



on 

i n two 

As S 

e 1 a t i o 

a bro 
3; the 
less b 
ver th 
ampl e 
parame 
1 1 ary 
he ran 
1 ar ex 

a r b i t 
= 0.21 

S r ha 
values 

on th 
1 es 7 

the 1 
vely ) . 
theref 
nd all 
s of T 

Sr = 
vely. 



cases 
r of s 
n coef 
ad abs 

decre 
al ance 
e rang 
12 (0. 
ters a 
rise p 
ge 0.1 
cept t 
rari ly 

i n Ta 
d rel a 

of Op 
e MRC 
and 12 
arger 

Thes 
ore co 

cor re 
a b 1 es 
0.21 a 



( samp 
ampl e 
f i c i e n 
ol ute 
a s e in 
d by a 
e S r = 
52 or 
nd coe 
ropert 

to 10 
hat no 

smal 1 
bl es 4 
t i vel y 

(Tabl 
betwee 

al so 
val ues 
e 1 a rg 
n s i der 
1 at i on 
6 and 
n d S r 



1 es 7 a 
1 2 was 
t gradu 
maximum 

res i du 
n i n c r e 

0.52 t 
0.1 3) h 
f f i ci en 
i es wer 

mm/day 

m a x i m u 

val ue 
, 5, 6 

little 
e 4), f 
n 1 and 
fall be 

of S r 
er val u 
ed pref 
s i n v o 1 
7 were 
= 0.52 



nd 12) 
decreas 
ally ch 

of -0. 
al s of 
a s e in 
o S r = 
ad some 
ts (Tab 
e rel at 

(Z of 
m was r 
of S r t 
and 7. 

impact 
or samp 

1 5 bar 
tween 1 
are use 
es of S 
era bl e 
v i n g t h 
entered 
for sam 



no cl e 
ed fro 
anged 
9979 c 
some p 
r e s i d u 
0.13. 

ef f ec 
1 es 4, 
i vely 
Table 
eached 
hat wa 

Once 

on th 
1 es ot 
. The 

and 1 
d (0.2 
r for 
to the 
e a', 

with 
pies 7 



ar ma 
m 0.5 
from 
enter 
oi nts 
al s o 

The 
t on 

5, 6 
unaf f 
7). 
» Sp 
s com 
again 
e Z o 
her t 

val u 
5 bar 
1 and 
sampl 

smal 
n ' , a 
these 

and 



ximum was 
2 to 
-0.9978 
ed on 

bei ng 
f other 
value of 
the 

and 7) 
ected 
Sample 7 
= 0.1 
pared 

the 
f Table 
han 7 and 
es of Op 

'o.52 
es 7 and 
1 er 
" and n" 

vari ates 
12 



The pore-size distribution index (A; Table 4) tended to 
decrease as percent sand decreased with sample 99 being a notable 
exception. Note that A is sensitive to values of S r (Table 4, 



- 17 - 



CD 




sz 




-M 




O 




-t-> 




co 




+-> 




c 




<o 




4-> 




(/) 




cr 




o 




o 




O) 




co 




CD 




4-> 








4- 




O4- 




CO 


-M 


CD 


■1— 


+-> 


4- 


•I— 




CO 


CD 




JZ 





■M 


«— 


■O 


E 


cr 


O 


<o 


c_ 




4- 


r< 


co 




CD 


-O 


^- 


c 


Q. 


<c 


E 




03 


-Q 00 


.cr 






CD I 


A 


C_ 


c_ 


00 







4- 


CO 





+-> 




cr 


CD 


m 


> 


-t-> 


(_ 


CO 


3 


c 


<J 










cr 







>> 


■1- 


CD 


-M 


C_ 


c: 


O 


CD 


c_> 


+J 




CD 


■0 


£_ 


cr 




03 


CD 




£_ 


CO 


13 


_*r 


-t-> 


O 


CO 


O 


•1— 


t_ 





CO 


E 


• 




*3- 




CD 




1 — 




jQ 




<n 





<< o 



4- 


T3 





cr 




03 


00 




CD 


4- 


4-> 




03 




E 




•r— 




+J 




CO 




LlJlch 



-Q E 



cr 

O C 

+-> cr 

03 CD "O 

r- 0) C 

CJ 5 (O 

C- 4-> 

S_ CD co 
O -Q UJ 
O 

cr 



"O 
CD 

13 
CD 
O 



CD 



CD 

a. o 

E cr 

03 
CO 



1 — WO 
«3" *d" LO 



OOO 



r— O O 



ro ro >* 



r^ r— ^r 

ro com 

O O CD 
CD CD O 



CD 

1 — «3" CO 

, — eg , — 
CD CD CD 



CT> 



CO 



CO 
LO 



"tMMTlO r— 

CMM. — CM «3" , — 



O CD O O CD 



CO LO CT) 


1 — 


CO CO CO 


■=d- 


• • • 


• 


OOO 






N IDfO N CM 



OOOOO 



o o 



OOOOO 



^f- CO 



c\j c\j co <3- co 



O O 



O O o o o 



o o 



O O CD CD CD 



LO Cn LO O CO 

LO CT> 1— • 1 — 1 



Cn 
O 











CO 


CM 


<d" CO 


LO ,— 


LO LO CO «3" CM 


CO «d" 


1 — 


LO CM 


<o 


lo lo r-^ co 00 


co r^. 


LO 


LO LO 


LO LO 


nroNcOr- 


CM LO 


1 


cn en 


LO CO 


«=J- , — CM CM CM 


t— 



o o 



Mi — LO 


r>. *3- 


CO 


LO "^" CO CT> 


LO 


co 


CO CM CO 


r— ^l- 


•— 


«* O CT> CTi 


LO 


00 


lo co r-» 


CM LO 


co 


•vj" r— CD CO 


CO 


LO 


1 — , — CM 


CM , — 




r— CM 







1 — 


r~ lo 


^i- lo 


LO 


CM CO CO 


LO 


co en 


Cn 


lo r-- 


*d- *3- 


<=d- 


co co en 


en 


r-» r-» 


cr, 


en en 


cn en 


en 


en en en 


en 


en en 


en 


Cn CTv 


en en 


en 


en en en 


en 


en en 






1 1 



1 1 


CD 

1 


OOO 

1 1 1 


CD 

1 



1 1 



«=S- "==r 



lolo cncocnocM 1 — lo 
r-^co roN<*OLfi 00 *3- 

OO 1 — Of—.— . — 1 — O 



O O 



LO LO 

, — O CO NlOOOi — CMCO 

CM, — «3" CM "3" CVJ LO LOi — 



O O 



-a 
cr 
n3 



cr 
o 






CD 

■o 



LO 



cr 

UJ 

CD 
CD 
CO 



CD 

t_ 
CD 

O 
4-> 

"3 E 

JC O 
4-> 

O 

CD CM 
> 

t- -o 

r3 c 

O 03 

cr o 

O r— 

+J « 

C LO 

CD 

-M " 

CD O 

t_ 

CD 

cd cr 

i~ -r- 

ZS CD 

CO 

•1— CO 

O +-> 



CD O 

_CT Q. 

c_ 
4- rs 
o o 

4- 
-o 

C +-> 
CD CO 



CO CD 

cr co 

CD CD 

+-> .cr 
-t-> 

O '" 

r— en 

CD • 

.cr cr 

+J UJ 

4-> c 

03 O 

CO CO 

-t-> CO 

cr cd 

•r- l_ 

O en 
a. CD 

CO 

+-> E 

03 o 

-a c_ 



o -o 

CD 

C_ "D • 

<U 3 L. 

jQ r— CD 

E O +-> 

Z5 X lO 

z: cd 2 



en 
00 



to 

cr 



03 
O 
CO 
en 



cr 

CD 

E 
CD 
O 

CO 

cr 
o 

+-> 

rs 

-Q 

t_ 
-M 
CO 



CD 
Nl 

CO 

CD 

U 



<T3 

en. 



CM 



c 

<T3 



03 
CO 



r— en 



T3 


•> 


c 


00 


03 






• 


03 


CO 


CO 


cr 



03 
o 

00 

en 



cr 

CD 



CD 
O 



CO 



cr 
o 

t_ 
03 
O 



T3 

cr 
03 

cr 
o 



co 



CD 

NJ 

CO 
CD 
O 



ra 

en. 



E 
o 



- 18 



samples 7 and 12); increase in S r leads to greater decrease in In 
E s over a given span of h above ht> since E s = 1 at h = h^ and, 
for a value of corrsponding to a given value of h > hb, 
increase in S r leads to decrease in E s . 

Note that the value selected for S r , and therefore the value 



taken by X, also influences the value taken by n of Eq. 9 
(Table 5). Because Eq. 8 is an approximate description of 
K:h relationship of Eq. 9 one can say, 



the 



K s h b (2 + 3X) /n (2 + 3X) a a/ ( n n + b ) and> 



(17) 



n - 2 + 3X 



Estimates of X obtained from particle size distribution . 
alone (Bloemen 1980a, Eqs. 8, 9, 10, 3a and 11) and particle size 
distribution plus organic matter (Bloemen 1980a, Eqs. 8, 9, 10, 
3a and 12) are included in Table 4 for comparison. Organic 
matter was assumed to contain 58% carbon (Hesse 1971, p 209). 
Agreement between these estimates of X and X obtained from the 
MRC was not close. 



with 



The X values 
texture than 



of Table 4 (obtained from MRC data) varied more 
would be expected on the basis of Brakensiek et 



al. (1981, Table 3) and the r values of Table 4 were larger than 
the values of residual saturation on a fractional bulk volume 
basis reported by Brakensiek et al. for corresponding textures 
(1 981 , O r of Table 3). 

The constants a, n and b of Eqs. 9 and 10 accounted for more 
than 99% of the variation in In K, (Table 5), where K was 
generated for the 12 points of the moisture retention curve using 
X, S r and K s in Eqs. 4 and 7, showing that Eq. 9 is an adequate 
representation of the Brooks and Corey (1964) prediction of the 
K:h relationships for these samples. The exponent n tended to 
decrease with decrease in percent sand, sample 99 again being a 
prominent exception. 

Entering Eq. 11 with K from Eq. 9 generates the height (Z, 
cm) above a stable water table that is consistent with some 
assumed steady state upward flux (q, cm/day or mm/day) from the 
water table when soil moisture tension at height Z is constant 



and is equal to h 



ma x 



(Fig. 4). 



represents the Z:h relation at 



The dashed 
q = 0. 



line of Fig. 4 



The regression equations of 
relationships between Z and q at 



Tables 6 and 7 describe the 



two values of h 



at three values of q (Table 
are included for comparison 



ma x • 



Values of Z 



7), as obtained directly from Eq. 11 
with Z that can be generated using a 



- 19 



E 
u 


160 

120 

80 

40 






y 


* 

/ 

i 
i 
i 

1/ 




1 ' T 1 

(mm/d< "' ) oio ; 

032 


X 

t- 
a. 


100 


Q 






316 








I0.Q 










316. 








IOO 






f . i . 1 . i . i i 



100 200 300 400 

TENSION (h.cm woter) 



500 



Fig. 4. Height (Z) above a stable water table and soil moisture 
tension (h) for selected values of steady state upward flux (q); 
sample no. 55. 



and n ' or a" and n". For example, in Table 7 In Z for q = 1 
mm /day (col. 5) is immediately comparable to a" (col. 2). The 
regressions (Tables 6 and 7) are almost exactly linear within the 
range of q under consideration (0.01 to 1 c m / d a y ) . As noted in 
Materials and Methods, the coefficient n ' of E q . 15 (Table 6) 
approximates the exponent n of Eq. 9 ( n ' - - n). Note that 
these constants (a 1 , n', a", n") do not apply beyond the range of 
q to which they are fitted. For example, compare a' for h 
4000 and h 
at Z = 1 ; 
of course 



max = 500 (Tabl e 6) , a ' being, 
a ' ( h m a x = 500) is larger than 
can not be correct. Because the 



max " 
in a literal sen se , In 
a' (h max = 4000) which 



i 



coefficients lead 



to q much greater than 1 
considered in isolation. 



when Z = 1 they are misleading when 



Relations between Z and q at h 



ma x 



= 500 cm water are more 



relevant to water supply and root function than at h max > x 50 0, 
500 cm water being roughly the point at which root growth becomes 
limited by water supply (Russell 1977, p 99). For the fitting of 
a linear In q : 1 n Z relationship a higher h max (e.g. 4000 cm 
water) was mathematically more desirable because the In q:ln Z 
relationship was then more nearly linear (Table 6). The 



curvature at lower values of h 



Z could not exceed h 



ma x 



arose 



in a x 



no 



latter how si 



from the constraint that 
1a 1 1 q wa s . 



Because the correlation between In q and In Z a p pro aches 
unity these two regression equations (Eqs. 15 and 16) approximate 



- 20 - 



Table 5. The constants a, n and b 
samples from 10 sites'*" 



of Eqs. 9 and 10 for core 



Sampl e 
no. 



Sum of squares 
of In K in 
regressi on 
(%) 



9 

11 

8 



55 
99 
5 
10 
13 

12 



0.1019 

0.234 

0.163 

0.21 
0.10 

0.435 

0.275 

0.46 

0.28 

0.51 



52 
13 



11 .84x1 6 
1 .477xl0 6 
2.11 3xl0 6 

9898 
1 346 

1 58.4 
0.1 756 

27.79 
124.8 
0.9143 

0.2910 
0.1 235 



5.529 
5.034 
4.919 

3.667 
3.275 

3.375 
2.394 
2.872 
2.862 
2.589 

2.218 
2.036 



9.608xl0 6 
3.635xl0 6 
21 .58xl0 6 

148670 
19193 

1 356 
40.84 
1071 
7722 
386.2 

1 26.5 
46.90 



99.8 
99.8 
99.7 

99.3 
99.5 

99.3 
99.7 
99.9 
99.8 

99.4 

99.2 
99.3 



t 



Obtai ned 
(Table 3 
Table 4, 



by regression 
0-500 cm) ; K 
the of the 1 2 



of In K on h at the 
was generated using 

points of the MRC and K s (Table 3). 



12 points of the 
the constants of 



MRC 



algebraic equations and accordingly 
n" - l/n\ 
a" - -(a' + In 1 ) /n ' and 



(18) 
(19) 



plots of q on Z , generated using Eq. 15, are similar to plots of 
Z on q generated using Eq. 16. 

As noted at the beginning of this section the hydraulic 
properties of the samples from the capillary rise sites (samples 
5 and 55) are not unusually conducive to capillary rise and fall 
well within the range of samples from the other eight sites 
(Table 7). Thus the substantial capillary rise that was observed 
in the experimental orchard (Table 1) would probably be equalled 
or exceeded on many soils of the region, given comparable 
conditions of water table depth. 



- 21 - 



00 

Cu 

to 

> 



■a 
cr 
<o 

00 

aj 



0+. 



E 


U 


O 




t_ 




4- 


l\l 


00 


cr 


CU 


1 — 



Q. C 



T3 


+ 


00 




OJ 


<T3 


L. 




O 


II 


o 






-^ 


t_ 


>■> 


o 


fC 


14- 


T3 



c o 



T3 




C 


cr 


<a 






cr 


— 


i — 


03 






CD 


oo 


t_ 


J-> 


ai 


C 


.cr 


T3 


2 


4-> 




00 


X 


c 


ro 


O 


e 


CJ 


-C 



CO 



X2 
03 



03 



o 
o 

LO 



X 

13 



CO 

CD cr 

e o 

03 t- 

rs oo 

O" 00 
oo CD ' 

i- ; 
i*- cjv 
o ai 

c_ 
E 

00 -r- 



+-> 

03 

3 



o 
o 
o 
■=3- 



00 

CD c 

i- o 

03 -i- 

13 00 

cr oo 

+- C0- 
O CD 

t_ 
E 

13 C 
OO -r- 



T3 
CT 
03 



O 



Cn LO 








en en co 


(£5 r— 


m^-i — r~- «vf 


CM CO 


CT> Cn Cn 


CT> O 


oin irnoo 


Cn r- 


cn cn cn 


CO CO 


CTi CTi CTi CO CO 


r^ CO 


cn cn cn 


CT> CTi 


CT> CTi <T> Cn CT) 


CT> CTi 


CTi Cn CT> 


cn cn 


CT> CTi CTi CTi CTi 


en cn 



OO «3" 0O 

lo o en 


en cn 

00 LO 


i — «=j- r-~ r^ ■vj- 

^- 0O CTi i — <3- 


CO CO 

r— O 


1 1 1 


oo oo 
1 1 


0O CM CM 0O CM 

1 1 1 1 1 


CM CM 

1 1 



oo en oo 


O CO 


oo cn co co lo 


i — l~~ 


CO CO CM 


O ■=*■ 


CTi CTi IT) O , — 


LO Cn 


OO r— CM 


CO CO 


CM LO , — «3" 1 — 


CO LO 


CM CM CM 









CTi 




CO 




CT> CTi 


en en 


cruo nm i — 


O CO 


cn cn 


en en 


en lo en en lo 


o oo 


cr. en 


en en 


en en en en co 


en en 


:> en cn 


en en 


en en en en en 


en en 


d cn en 


en en 


en en en en en 


en en 



ro 


ro 


en 


LO 


co 


r--. , — 


LO 


ro 


1 — 


CD CO 


LO 


o 


CO 


CO 


CM 


oo oo 


CO 


CO 


"=3- 


i— en 


LO 


1 


1 


oo 
i 


OO 

1 


0O CM 

1 1 


CVI 

1 


CM 

1 


CM 

1 


CM i— 

1 1 



locolo *d-oo o o rv o >* coco 

cor--o cocn cocnoLOO cmco 

OO.— 'CM CO^t" CMLOi — M N COLO 

CMCMCM r- t— •— i — .— 

, — colo lolo r^ro^r«3-co lolo 

ro cm . — <3- «^- cor^cOLO<3- <3- ^3- 

encnen coco r-- co co lo lo ^-^j- 



co 



.— 1 












cm ro 


CM , — 












LO , — 


• • 












• • 


II II 












ii ii 


t_ t_ 












l_ £_ 


oo oo 












OO OO 


r-- 


LO 


ai 


LD 


a 


CO 


CM 




LO 


en 




, — 


. — 1 


. — l 





X 


>^ 


<o 


-Q 


E 




-c 


"O 




O) 


ii 


4-> 




ro 


-CZ 


i_ 




a> 


T3 


cr 


c 


CU 


<a 


CD 






o 



CM II 

• 

CD -C 
O 

. — C 
CD 

•r- a> 
cr 5 

-t-> 
II CD 

-O 

Too 



cr 


e 




cu 


sr 


E 


+J 


CD 


■<— 


t_ 


S 


O 




cr 


">, 


■i- 


03 




"O 


cr 


">>^ 


o 


E 


>r- 


O 


oo 




cr 


o 


CD 


• 


■M 


1 — 






o 


o 


o 


+-> 


LD 


, 


sz 


o 


4~> 


• 


•1— 


o 


2 



o <— 

cr . 
cr 

£_ LiJ 

o 



M 


cr 




o 


4- 


•r- 


O 


+-> 




n3 


oo 


t_ 


cu 


en 


r3 


CD 


i — 


+-> 


n3 


cr 



- 22 - 



o 
o 
o 



X 

E 






to 

<D 
+-> 
•i — 

CO 

O 



E 
O 



00 
CD 



(/) 

<D 

o 
o 

£_ 
o 
it- 



er 
-o 
to 



oo 




+J 




C 




<o 




■M 




oo 




c 




o 




c_> 


CD 




c_ 




CD 


• 


.C 


r-^ 


s 


cu 


C 


i — 


CD 


-Q 


■M 


<TD 


13 









CO 

CD 



A3 



E 

C 

>> 

-Q 

"O 
CD 
C 

•r— 
fO 

+-> 

.O 
O 

CO 
fO 



cr 






i — 


c 


• 


'— 


o 


c 


"O 




c: 


+ 


13 


= 


, — 


13 






n 


II 


O 




r— 


E 


II 


o 




1 — ' 


cr 


r^-j 


-t-> 




13 


c 





l^-J 



oo 




CD 


i — 


rs 


■— 


13 


• 


> 


cr 




LiJ 


cr 


<4- 


13 


o 



r-j 



>> 

-o 

E 
E 



o 
II 

cr 



>> 

A3 

-a 



ii 

cr 



13 

-a 



n 

Cr 



ai 
a. o 

E C 
13 

oo 



in en oo lo en n en *d- cm cm 

, — i— I CM COCO r— CM «tf- . — 



CM LO 



«d" O LO 
r- CM >vf 



o r--~ 
en en 



Cn LO o oo o 

00 0O .— CO LO 



o o 

LO LO 



LO LO O 

r-^ r-» en 



o r^ 
o cr. 



LO oo 00 CM oo 



cn r~~ 



r— en -^- *3- lo 

oo en o r-~ O 

.— .— CM CM ro 



rvrOi — co lo eo o 
en oo lo lo i — i — i — 

CM «3" OO 0O ^" <d" LO 



o o o 



o o 



CD O O O O 

I I I I I 



o o 



ooenco «3-oo oOLoenoooo con 
p~.r^en cmcm <3- lo lo cm oo oo 



<3- <?r «d- 



LO LO 



«3- oo «3- m oo 



^t- <3- 



cn .— i oo 



r— 












CM OO 


CM ,— 












LO ,— 


• • 












• • 


II II 












II II 


L_ L_ 












t_ {_ 


oo oo 












OO oo 


r^ 


LO 


en 


LO 


o 


oo 


CM 




LO 


en 




i — 


, — 


i — 



- 23 - 



Comparison of observed with calculated capillary rise 

Comparison of observed rise with rise calculated from core 
properties (entering Eqs. 9 and 11 with the constants a, n and b for 
samples 55 and 5, with h max equal to soil moisture tension in the 
open cylinder at a depth of 40 cm and Z equal to water table depth 
minus 40 cm) provides a crude estimate of the quantative accuracy of 
the capillary rise projections of Tables 6 and 7. Calculated and 
observed rise for site #1 were in good agreement (Table 8). Rise 
calculated for site #2 was up to eight times as large as observed 
rise. 



Table 8. Comparison of observed capillary rise at two sites with 
rise calculated by numerical integration of Eq. 11 for the fie.ld 
conditions of daily Z (water table depth minus 40 cm) and daily 



'max 



(soil moisture tension at 40 cm) 



Locat i on 
Peri od 



Observed rise 
(cm) 



Cal cul ated 
(cm) 



n se 



Si te 1 (sampl e 55 ) 
4 Aug-1 Sept, 1 957 
21 June-5 July, 1 978 
4 July-26 July, 1 979 
27 July-3 Sept, 1979 

Site 2 (sample 5) 

9 Aug-1 Sept, 1 977 
21 June-5 July, 1 978 

20 July-20 Aug, 1 979 

21 Aug-9 Sept, 1 979 



4.3 (2.6-6.0) 

3.3 (1 .7-4.9) 
5.0 (3.5-6.5) 



4.4 (2.8-6.0) 

4.7 (3.7-5.7) 
2.7 (0.7-3.7) 



f 



3.6 (2.5-5.1 )* 
0.1 (0.05-0.2) 
1.5(1 .0-2.4) 
2.9 (1 .9-4.4) 



26.7 (8.0-88.9) 
1.6 (0.3-8.2) 
37.4 (1 0.6-1 31 .3) 
10.1 (2.4-42.5) 



t 



95% confidence range, from Table 1. 

*95% confidence range based on rise calculated from each of four 
cores ( see text ) . 



The calculated rise values of Table 8 were derived from mean 
K s (geometric) and mean MRC (arithmetic) of the four cores per 
site as described in Materials and Methods. The 95% confidence 
ranges of these estimates, obtained as described below, show that 
the cores from site #2 were much more variable than the cores 
from site #1. These confidence ranges were obtained by calculating 
rise for each of the four cores per site. 



- 24 - 



95% confidence range = C/F to CF 

where C is rise calculated from pooled values of K s and MRC, 

F = exp (3.182 SD/2) and 

SD is the standard deviation of the natural logarithm of rise 
calculated from each of the four cores. 

Whether the high variability of cores from site #2 reflects 
inherent variability within the undisturbed profile or 
modification of some core samples during sample extraction and 
processing is an open question. Whatever the cause of the high 
variability, it accounts in part for the discrepancy between 
observed and calculated rise on this site. Some possible causes 
of the residual disagreement between observed and calculated rise 
are explored in Appendix A. After adjustment for one of these 
causes (Appendix A) mean calculated rise on site #2 was about 
four times as large as observed rise and had 95% confidence 
limits that enclosed observed rise. 

This four-fold discrepancy and the wide confidence ranges 
(Table 8) serve notice that the capillary rise projections of 
Tables 6 and 7 may be rather poor predictors of rise in the 
context of the intact source profile and show that some aspect of 
the method, perhaps soil modification during collection of cores, 
was unsatisfactory. Note however that these projections (Tables 
6 and 7) may have a much smaller error when applied in the 
context of uniform profiles with properties identical to those of 
the core samples after collection. A four-fold discrepancy 
compares favourably with the error of ± two orders of magnitude 
discussed by Stallman and Reed (1969). 

Correlations of hydraulic properties and derived parameters with 
other soil properties 



The va 
deri ve 
proper 
struct 
space . 
cont i n 
quant i 
n " i n d 
soils 
result 
on por 
change 
betwee 



lues 
d pa 
ties 
u re , 

uous 
t i es 
epen 
n" a 
ed f 
e si 
s i n 
n A 



of hy 
ramete 
such 
etc. 
depen 
pores 
of la 
dent o 
nd In 
rom t h 
ze was 

pore 
and n" 



draul 
rs , s 
as de 

Both 
ds mo 
, whe 
rge a 
f K s . 
K s ar 
e f ac 

si mi 
size 

can 



i c pro 
uch as 
nsity , 
K s an 
st cr i 
reas n 
nd sma 
Tabl 
e high 
t that 
1 ar an 
d i s t r i 
be see 



perties (K s in this case) and the 
a" and n" depend on other soil 
texture, pore size distribution, 
d n" relate to the nature of the pore 
tically on the size of the largest 
" is a measure of the relative 
11 pores. Thus it is possible to have 
e 9 shows, however, that for these 
ly correlated. This high correlation 
in these sandy soils the upper limit 
d differences in K s resulted from 
but ion A or n". The direct connection 
n from the following approximations: 



- 25 - 

n - 2 + 3X (Eq. 17) 
n' - -n (Tables 5 and 6) and 
n" -1/n' (Eq. 18) 
from which n" - -1/(2 + 3X) 



(22). 



{For the samples studied here the correlation coefficient 
between n" and -1/(2 + 3X ) was 0.99 and n" = 0.042 + 1.17 
(-1/(2 + 3X) confirming, at least for these samples, that n 
and X are closely related and Eq. 22 is approximately 
correct. } 



both 

high 

betw 

met h 

rel a 

acco 

of v 

thes 

cont 

curv 

st ru 

was 

20 a 

samp 

Eqs . 



abo v 
of 1 
cor r 
betw 
and , 
a po 
i nte 

Db a 

posi 

dens 

to 1 

103 

sugg 

whi c 

betw 

dens 



Table 9 s 
highly p o 
ly negat i v 
een 1 n K s 
odol ogi cal 
t i v e 1 y con 
unted for 
a r i a t i o n i 
e soils, s 
rol s of 1 n 
e ) . The i 
cture and 
apparent ly 
nd 21 may 
1 es under 
20 and 21 



The 
e a 

mm/ 
el at 
een 

eve 
or e 
ract 
nd a 
t i ve 
ity 
.5, 
cm. 
est 
h a" 
een 
i t i e 



param 
wat e r 
day ca 
e d wit 
a " and 
n for 
st i mat 

to in 
" was 

and t 
of the 
exp a" 

The r 
that t 

i s ma 
a" and 
s abov 



hows t 
s i t i v e 
ely co 
and ot 

bias 
stant 
9 3.7% 
n n" ( 
and co 

K s an 
nf 1 uen 
part i c 

smal 1 
be rea 
con s i d 

can n 

eter a 
table 
n be s 
h bul k 

Db ac 
soils 
or of 
f 1 uenc 
shown 
hen ne 
i r s a n 

i ncre 
esul ts 
he cri 
x i mum , 

D5 wh 
e t he 



hat 1 
ly co 
rrel a 
her s 
in K s 
acros 
of th 
Eq. 2 
ntent 
d n" 
ce of 
1 e si 
. Es 
sonab 
e r a t i 
ot ap 



n K s a 
rrel at 
ted w i 
oil p r 
usual 
s samp 
e van' 

1). E 
and b 
(the s 
other 
z e d i s 
t i mate 
ly cor 
on but 
ply to 



nd der 
ed wi t 
th Db. 
opert i 
ly was 
1 es . 
at i on 
q u a t i o 
ul k de 
hape o 
possi 
t ri but 
s of K 
rect f 
it is 
soils 



i ved 

h per 

The 

es su 

eith 

Perce 

in In 

ns 20 

nsi ty 

f the 

ble f 

ion w 

and 

so 

impo 

in g 



s 

or 



paramet 

cent sa 

s e high 

ggest t 

er smal 

nt sand 

K s (Eq 

and 21 

were t 

m o i s t u 

actors 

i t hi n t 

n" obt 

lis s i m 

rtant t 

eneral . 



er, n , were 
nd and were 

correl at i ons 
hat the 
1 or was 

and Db 
. 20) and 9 6.5% 

imply t hat , for 
he dominant 
re retention 
such as 

he sand fraction 
ained from Eqs. 
i 1 a r to the 
o stress that 



" (t 
or t 
usta 

den 
cou n 
very 
a", 
e th 
by B 
gat i 
dy 1 
ased 

of 
t i ca 

i s 
i ch 
c ri t 



he n 
he t 
i ned 
s i ty 
ted 

si m 

It 
e va 
oone 
ve a 
oam 

fro 
Boon 
1 D b 
cl OS 
we f 
i cal 



atu r 
hick 

wi t 

, D b 
for 
i 1 a r 
i s p 
1 ue 

et 
s so 
Ap 2 
m 33 
e et 

i n 
e to 
ound 

va 1 



al lo 
ness 

n h ma 
. Th 
onl y 

to t 
robab 
of a" 
al. ( 
i 1 wa 
h o r i z 

to 1 

al . 
sandy 

1 .4. 

i ndi 
ue . 



gari thm 
of soi 1 
x = 400 
e regre 
64.3% o 
hose un 
le that 
. The 
1978, F 
s compa 
on was 
44 cm a 
(1978, 
1 oam a 
The n 
cates t 



of t 
aero 
cm ) 
s s i o n 
f the 
der c 
text 
rel at 
i g . 8 
cted . 
i nc re 
nd th 
Figs, 
nd 1 o 
e g a t i 
hat t 



he h e i 

s s whi 

was n 

rel at 

v a r i a 

o n s i d e 

u re an 

i onshi 

) to b 

As t 

ased f 

en dec 

8 and 

amy sa 

ve cor 

hese s 



ght, Z, 
ch a flux 
egati vely 
i o n s h i p 
t i o n in a " 
ration, is 
d dens i ty 
p between 
e f i rst 
he bulk 
rom 1.33 
reased to 

9) 
nd , at 
relation 
ampl es had 



- 26 - 



to 
a; 

•I— 

+-> 

t_ 

CD 
Q. 
O 



O 
00 



CD 



4-> 






O 






T3 


CD 




C 


£_ 




03 


CD 




CO 


S 




a> 






•r* 


A 




+-> 


CO 




t_ 


c -o 




cd 


o c 




a. 


•r- fQ 




o 


+-> 




£_ 


TO -— 


' — * 


CL 


=3 E 


>> 




cr a 


03 


O 


a; — 


■o 


•r— 




~~^ 


i — 


C M 


E 


^ 


O 


E 


03 


•r- c 


» — ' 


£_ 


00 i — 




"O 


00 


cr 


>~> 


CD - 




jz 


i- c 


c 




CO 


i — 


c 


cd + 




cd 


t_ 


= 


<d 


- 


c 


s 


■O 03 




4-> 


£Z 


+ 


CD 


0D II 




JO 




= 




00 ' — - 


03 


CO 


+-> >1 




Q. 


c: 03 


II 


•r— 


cd "a 




-C 


•i— \ 


- — » 


GO 


u E 


E 


c 


•r- a 


u 


o 


4- 


» — ' 


■r- 


4- 




+-> 


a> cri^i 


to 


o 




1 — 


o C 


c 


a; 


i — 


1 — 


C£ 


cz 
o 

•1 — 




• 


-t-> 




CT> 


13 





00 
-M 

ai 



03 



CD 



O 



CD cd 

I — L_ 

JO L_ 

03 O 

I— O 



lo 

• 

. o 





* 




* 




* 


LO 


r-~ 


LO 


cn 


• 


• 


o 


o 

i 



* * 




00 


* * 




+-) 


* * * 




c 


co cr> oo 


- 


a> 


cr> lo en 


03 


•i— 



o o o 



* * * 




* * * 




* * * 


t/J 


LO LO CM LO 


iZ 


CftCTllDCTi 





o o o o 



***** 
***** 
«3- r»» , — o i— 

CO 00 CO CO 00 



o o o o o 
II II 





* 


* * 


* 






-X 


* * 


* 






-X 


* * 


* 




1 — 


o 


O CM 


r-. **• 


-a 


co 


<n 


cn cn 


■=3- cn 


c: 

03 


o 


o 


o o 


o o 


UO 



o 



03 

CD 
C_ 

o 



JO 

Q 



jo 

Q 



i — r-^(D N't 
LO LO LO LO «^- CO 



o o o o o o 
I I I 



03 

Cl 
CD 



* 
* * * * 
i — r-^ i — oo cm 10 i — 
^t m <t ^o n i^iv 
• •••••• 

o o o o o o o 
t I I 



* 
* * * * 
r^ lo co lo lo co o 

«3- «3" <3" LO LO LO 00 

o o o o o o o 
I I I 



cj 

03 



c E 



o-? 


o 


00 










"O 





i^ 










C 














13 


JO 


c: 




- 


= 


= 


CO 


Q 


i — 


13 


£Z 


13 


C 







E LO 






LO O 






.— o • 






• • o 






O O 1 


-a 


+J 


>, 1 1 LO 


c: 


i — 


03 LO LO CM 


03 


■r— 


i — • • • _Q 


LO 


00 


uoooo 



cr 




o 




1 — 


O 


00 


CM 


00 




CD 


• 


£_ 


cr 


CD 


LU 


CD 





1/3 

CD 

£_ 
03 

rs 

ol 

co 



00 

c 
o 

-l-J 

03 

cr 
CD 

c 
o 
•i- 
co 
co 
CD 
c_ 

CD 
CD 



CM 



CO 

CM 





1 — 1 


o 




1 — 1 


, — 


V 




1 — 


o 


Q_ 




o 


o 






o 


• 


*i 




• 


o 


5^ 




o 


V 


CO 




V 


O- 


• 




o_ 




o 






m 


°o 




«\ 


&s 






s-s 


LO 


II 




r-. 


• 


CO 




• 


CO 


CO 




CO 


CT> 






CTi 




A 






II 


r-- 




II 


CO 


CM 




CO 


CO 






00 




ii 






A 


= 




n 


CM 


03 




i — 


CM 






LO 


o 


to. 




• 


• 


X 




o 


o 


CD 
Q 




II 


II 


CO 






£Z 


1 1 




00 


Q 






^ 


CO 


JO 




c 


1 1 


Q 




i — 




00 




Q 


JO 


o 




CO 

1 1 


Q 
CM 

o 


CM 
1 




JO 


CO 


, — ■« 




o 


• 


E 




vo 


o 


E 




«* 


1 






• 




LO 


• 


r-~ 


-a 


o 


i — 


i 


c 


• 


o 




03 


o 


o 


"O 


00 


1 


• 


c 




LO 


o 


03 


s-s 


CM 


V 


00 


1^ 


• 


o_ 




i — 


o 


* 


<?« 


"^J- 




* 


CM 


o 


^1 


* 


«* 


o 


— 




r^ 


• 


cr> 


• #> 


O 


o 


o 


1 — 


• 


+ 


• 


o 


CO 




CM 


• 


+ 


LO 


+ 


o 




a> 




V 


00 


o 


00 


Q_ 


LO 


• 


cn 


* 


• 


o 


CO 


* 


<3- 


i 










ii 




II 


it 




LO 






= 


O 


oo 




03 


• 


^ 






o 






Q_ 


V 


d 


= 


X 


Q- 


1 — 


c 


a> 


+ 



27 - 



Although the total quantity of sand was not correlated with 
a", the three sand fractions each had a positive correlation with 
a" and exp a" (Table 9). The correlation between exp a" and the 
0.05 - 0.25 mm sand fraction was slightly better than with other 
soil fractions. A regression equation of exp a" (Eq. 23, Table 
9) against the 0.05 - 0.25 mm fraction and Db accounted for 80.3% 

in exp a". This tendency for exp a" to increase 
in fine and very fine sand content is consistent 
to maintain a given constant flux over greater 
depths of fine sand than for coarser sand (Bloemen 1980b). 



of the variation 
with an increase 
with the ability 



In the sandy soils studied here their high density, fine 
sand and coarse sand contents have been shown to be the dominant 
properties which influence the capillary rise properties. Ranges 
of density and sand contents can be used to group or classify, 
sandy subsoils. The probable influence of the dominant 
properties on capillary rise can be estimated using the framework 
provided by Eqs. 20, 21 and 23. 



For ease of visualizing the effect of the dominant 
properties some relevant data from other tables are drawn 
together in Table 10. The soils where high density, fine sand or 
coarse sand dominate the capillary rise are designated D, FS and 
CS, respectively. A fourth class, I, came out where none of 
these properties were dominant and the resultant capillary rise 
indicators had intermediate values. It is important to stress 
that these classes of soils are not distinct but only points on a 
continuum. The capillary rise indicators are exp a" (the depth 
of soil over which a flux of 1 mm /day can be maintained by a 
constant tension difference of 4000 cm), n" (an indicator of pore 
size distribution) and AZ/Aq for two ranges of q (i.e. the change 
in water table height which leads to a ten fold change in flux 
q). 

D soils - 1) In soils with unusually high D D , exp a" is 
small, perhaps as small as 35 cm (Eq. 23, Table 10). 

2) The high D D results in small K s (Eq. 20), low 9 S and a 
tendency for large r . This means that there is low water 
yield from these soils (i.e. O s - O r is small) and the flow 
of water for replenishment is slow, as a result of the low 

Kc. 

3) The constant n" is small indicating a wide pore size 
distribution which interacts with the low water yield and 
and results in a great sensitivity of flux to changes in 
water table height e.g. AZ of 22 cm causes q to drop from 10 
to 1 mm/day. 

For a significant rise of 1 mm /day in dense sandy subsoils 
the water table must be shallow and stable but the low water 
yield and low rate of lateral down slope flow will tend to make 



- 28 - 



Table 10. Selected data and characteristics of the core samples 
arranged by type 



Type Predominant Sample Bulk Total Fine sand Coarse sand 
factor number density sand 0.25-0.05 mm 2.0-0.25 mm 

(9/cm3) (%) {%) (%)- 



D 


Density 


99 
13 


1.99 
1.88 


67.3 
54.6 


32.1 
25.3 


35.4 
29.3 


FS 


Fine sand 


10 
7 


1.67 
1 .67 


55.4 
84.5 


45.5 
69.1 


9.9 
15.4 



FS-CS 



1 .67 



91 .5 



48.5 



43.1 



CS Coarse sand 11 1.61 92.6 38.8 

9 1.56 93.1 23.1 

D-I 12 1.78 44.5 25.2 



53.8 
70.0 

1 9.3 



Intermediate 55 1.75 73.7 24.3 
5 1.83 66.4 23.6 



49.5 
42.7 



Table 10. (continued) 



- 29 - 



Type 



exp a 
(cm) 



Sensitivity AZ/Aq Water yield 

(Zl-Zio) § (Zn.i-Zi)S (O s -O r ) ■ 
(cm ) (cm ) 



D 


35 
48 


[51 ] + 
[60] 


-0.43 
-0.42 


[-0.42]* 
[-0.44] 


22 
32 


59 

73 


0.206 
0.147 


FS 


187 
189 


[146] 
[195] 


-0.35 
-0.27 


[-0.37] 
[-0.25] 


106 
90 


232 
1 65 


0.257 
0.284 


FS-CS 


145 


[152] 


-0.20 


[-0.22] 


55 


86 


0.276 


CS 


120 
11 3 


[144] 
[122] 


-0.20 
-0.18 


[-0.20] 
[-0.18] 


44 
39 


70 
58 


0.267 
0.330 


D-I 


59 


[80] 


-0.48 


[-0.45] 


41 


112 


0.1 68 


I 


89 
109 


[85] 
[67] 


-0.30 
-0.35 


[-0.32] 
[-0.37] 


44 
62 


87 
134 


0.181 
0.1 75 



t 



23. 



Values in brackets are estimates of exp a" from Eq. 

^Values in brackets are estimates of n" from Eq. 21. 

§(Z-|-Zio) is the change in water table depth that will decrease q 
from 10 mm/day to 1 mm/day and (Zq- 1 -Zi ) is the change in Z that 
will further decrease q to 0.1 mm/day (Table 7); the smaller the 
change in Z for a decade change in q the greater the sensitivity 

.of q to Z . 

iNote that the rank of types is as follows: 



for exp a", FS > CS > I 
for n" , CS > FS = I > D 
resultant sensitivity of 
for q > 1 mm/day FS < I 
for q < 1 mm/day FS < I 



> D, 

q to Z i s 
- CS < D and 
< CS - D. 



30 - 



the water table unstable. Consequently, capillary rise of water 
in such subsoils is unlikely to make a significant contribution 
to the water supply for plant growth under field conditions. 

FS soils - 1) Soils with moderate Db and in which the fine 
sand fraction is a major component have large exp a", 
perhaps as large as 190 cm (Eq. 23, Table 10). 

2) The values of n" are intermediate and K s are also 
intermediate indicating that water removed may be 
replenished if an upslope source exists. 

3) The flux of water is relatively the least sensitive to 
changes in water table height. 

Of the soils considered this is the class of soils for which 
capillary rise is most likely to be a significant source of 
water. A rise of 1 mm/day can be sustained over depths 
approaching 2 m below the rooting zone and the flux is relatively 
insensitive to changes in the level of the water table. 



CS Soi 1 s - 1) Sandy soils in 
dominates tend to have lower 
a wide particle size distribution 
are intermediate between FS and D 

2) The constant n" is largest for 
the narrow pore size distribution 
sensitivity of flux to changes in 
40 cm drop in 
(Table 1 0). 

3) Val ues of K s 
under consideration. 



which the coarse sand fraction 
bulk density than do soils with 
The values for exp a" 
soils (Eq. 23, Table 10). 
these soils as a result of 

This leads to a high 
water table level e.g. a 
water table would reduce q from 10 to 1 mm/day 

will be maximum for the range of soils 



For capillary rise to be significant in these soils a 
relatively stable water table at intermediate depths is required. 

The samples labelled I in Table 10 represent soils that are 
intermediate between types D, FS and CS; sandy loams of moderate 
D5 without predominance of fine sand and the resultant capillary 
rise indicator parameters are intermediate in value. Those 
having values intermediate between two classes are so listed in 
Table 1 0. 

Influence of capillary rise on crop growth 

A shallow and sufficiently stable water table would be a distinct 
advantage in sandy soils that have a low plant available water 
content when fully drained. For example, the performance of 
apple trees was reported to be better on imperfectly drained 
sandy soils than on those which were well drained (Deckers 
1975). Tamasi (1964) reported better growth of 4-yr-old Jonathan 
apple trees in the portions of one orchard (sandy soils) with 



- 31 - 



shallow water table (50-60 cm spring, 80-90 cm Aug.) than in 
portions with deeper water table (150 cm spring, 220 cm Aug.). 
All trees were sprinkler irrigated five or more times during the 
growing season. Over the shallow water table the trees tended to 
lean and to develop chlorosis but weighed 2.4 times as. much as 
did trees on the better drained areas. Similarly, yields of 
irrigated corn and sugarbeet on sandy loam and loamy sand 
overlying sandy subsoil decreased as water table depth increased 
beyond the optimum depth of about 130 cm (Benz et al . 1981). 



The present study indicates that rise is a potential source 
of site to site and year to year variation in the water supply 
for crop growth in the Atlantic region. Some rapid and reliable 
means of estimating capillary rise under field conditions would 
be desi rabl e. 



32 - 



REFERENCES 

BENZ, L. C, DOERING, E. J. and REICHMAN, G. A. 1981. Water 
table management saves water and energy. Trans. ASAE . . 
24:995-1001. 

BLOEMEN, G. W. 1980a. Calculation of hydraulic conduct i vi t'i es of 
soils from texture and organic matter content. Z. 
Pflanzenernaehr. Bodenkd. 143:581-605. 

BLOEMEN, G. W. 1980b. Calculation of steady state capillary 
rise from the groundwater table in mu 1 t i -1 ayered soil profiles. 
Z. Pflanzenernaehr. Bodenkd. 143:701-719. 

BOONE, F. R., B0UMA, J. and de SMET, L. A. H. 1978. A case . 
study of the effect of soil compaction on potato growth in a 
loamy sand soil. 1. Physical measurements and rooting 
patterns. Neth. J. Agn'c. Sci. 26:405-420. 

B0UWER, H. and JACKSON, R. D. 1974. Determining soil 
properties. Pages 611-672 in J. van Schilfgaarde ed., Drainage 
for Agriculture. Agronomy No. 17, Amer. Soc. Agron., Madison, 
Wisconsin, U. S. A. 

BRAKENSIEK, D. L., ENGLEMAN, R. L. and RAWLS, W. J. 1981. 
Variation within texture classes of soil water parameters. 
Trans. ASAE 24:335-339. 



BROOKS, R. H. and COREY, A. T. 1964. Hydraulic properties of 
porous media. Colorado State Univ. Hydrology Papers No. 3, 27 
pp. 

CANN, D. B., MacDOUGALL, J. I. and HILCHEY, J. D. 1965. Soil 
survey of Kings County Nova Scotia. Nova Scotia Soil Survey 
Report No. 15. Can. Dept. Agric. and N. S. Dept. of Agric. and 
Ma rket i n g , 97 pp. 

DECKERS, J. C. 1975. Choix des especes fruit i e r e s selon 
l'humidite des sols. Le Fruit Beige 43:127-132. 

D00RENB0S, J. and KASSAM, A. H. 1979. Yield response to water. 
FA0 Irrigation and Drainage Paper No. 33, FA0, Rome, 193 pp. 

DRAPER, N. R. and SMITH, H. 1967. Applied regression analysis. 
John Wiley and Sons, N.Y. 407 pp. 

GARDNER, W. R. 1958. Some steady-state solutions of the 
unsaturated moisture flow equation with application to 
evaporation from a water table. Soil Sci. 85:228-232. 



- 33 - 



HESSE, P. R. 1971. A textbook of soil chemical analysis. John 
Murray Ltd., London, 502 pp. 

JAWORSKI, J. 1969. Evapot ranspi rat i on of plants and fluctuation 
of the groundwater table. Pages 730-739, in P. E. Rijtema and 
H. Wassink eds., Water in the unsaturated zone. Vol. 2. IASH, 
Ceuterick, Belgium, 995pp. 

KLUTE, A. 1965. Laboratory measurement of hydraulic 
conductivity of saturated soil. Pages 210-221 in C. A. Black et 
al., eds. Methods of soil analysis. Part 1, Agronomy No. 9, 
Amer. Soc. Agron., Madison, Wisconsin, U. S. A. 

MASON, D. D., LUTZ, J. F. and PETERSEN, R. G. 1957. Hydraulic 
conductivity as related to certain soil properties in a number of 
great soil groups - sampling errors involved. Soil Sci. Soc. 
Amer. Proc. 21 :554-560. 

McKEAGUE, J. A. ed. 1978. Manual on soil sampling and methods 
of analysis. Can. Soc. Soil Sci., 2nd edition, 212 pp. 

RICHARDS, L. A. 1965. Physical condition of water in soil. 
Pages 128-152 in C. A. Black et al . , eds. Methods of soil 
analysis. Part 1, Agronomy No. 9, Amer. Soc. Agron., Madison, 
Wisconsin, U. S. A. 

RUSSELL, R. S. 1977. Plant root systems. Their functions and 
interactions with the soil. McGraw-Hill, London, 298 pp. 

RUSSO, D. and BRESLER, E. 1980. Field determinations of soil 
hydraulic properties for statistical analysis. Soil Sci. Soc. 
Amer. J. 44:697-702. 



SWANS0N, C. L. W. 1950. A portable soil core sampler and 
penetrometer. Agron. J. 42:447-451. 

STALLMAN, R. W. and REED, J. E. 1969. Steady flow in the zone 
of aeration. Pages 565-578 in P. E. Rijtema and H. Wassink eds., 
Water in the unsaturated zone. Vol 2. IASH, Ceuterick, Belgium, 
995 pp. 

TAMASI, J. 1964. [The effect of the level of groundwater on the 
roots of apple trees] (Hung.) Maggar Tudomanyos Akademia 
Ag ra rtudomany i Osztalyi Kozlemeny 23:25-41. 

T0PP, G. C. and ZEBCHUK, W. 1979. The determination of 
soil-water desorption curves for soil cores. Can. J. Soil Sci. 
59:19-26. 

WEBSTER, D. H. 1978. Soil conditions associated with absence or 
sparse development of apple roots. Can. J. Plant Sci. 58:961-969 



- 34 - 



APPENDIX A 

Discrepancy between observed and calculated capillary rise at site 
#2 (sample 5) 

The disagreement between observed and calculated rise at site #2 
could have arisen in several ways. Some of these are as follows: 



1) 



The K s and MRC of sample 5 may be in error. Because S of 



sample 5 is larger than total porosity (Table 3) it is 
probable that the MRC is somewhat in error and K s of core 
samples, in particular if the soil is compact, has a large 
methodological uncertainty (as opposed to statistical 
uncertainty) arising from possible disturbance of soil during 
the process of core sample collection and preparation and from 
the unknown contribution of flow along the interface of core 
sample and core retaining cylinder to measured K s . 

2) The cores that constitute sample 5 may not be representative 
of the soil volume underlying the open cylinder. Some 
discrepancy between sample values and the true value of the 
volume sampled is to be expected. 

3) The method of generating K from K s and MRC may be 
inappropriate. In view of the possible errors under 1 and 2 
there is no reason to conclude, at this point, that the method 
of generating K was inappropriate. Adjustments to the exponent 
in Eq. 8 have been proposed such that 2+3X becomes 1.4+3A 
(Bloemen 1980a) or 2 + 2A (Russo and Bresler 1980) but these 
adjustments, applied to Eq. 7, would increase the disparity 
between observed and calculated rise. 



4) The calculation of expected rise from the field records, 
during a rapid decrease in water table depth and during the 
subsequent period of shallow water table may be inaccurate due 
to marked departures from steady state conditions. 
Examination of the field records and calculations on a daily 
basis (Table Al) indicates that much of the discrepancy arises 
during these periods. The following paragraphs are directed 
to considering and isolating this discrepancy arising under 
item 4 so as to obtain a better estimate of the discrepancy 
arising under items 1 and 2. 

First consider the increasing sensitivity of calculated 
steady state q to changes in soil moisture tension (h max ) and Z 
(water table depth minus 40 cm) as h max and Z decrease using 
sample 5 as the example (Table A2). Note that tensiometer gauge 
error ( ± 20 cm at equilibrium) will introduce large errors in q 
only when both h max and Z are small, e.g. at Z = 1 the effect of 
20 cm i nc rement s i n 



'ma x 



However at all values of h 



on q i s sma 11 when h 



max 



(excepting h 



max 



max 



> 70 cm. 
e -9° n max 



< Z + 1 ) 



- 35 - 



Table Al . 


Daily vc 


ilues of Z and h 


max a ^ S1 


te 2 


(1979), cumulative 




class A 


pan 


evaporati on 


at the Kentvi 


lie 


weather 




station 


and 


cumulative 


ri se as ca 1 cul 


ated 


1 from sample 5 




Z 




n max 


A 




Cal 


t 
cul ated ri se 




(cm) 


( 


cm water) 


(cm) 






(cm) 


July 20* 


51 .5 




95 


0.56 






0.61 


21 


55.1 




125 


1.14 






1.18 


22 


58.8 




1 55 


1 .62 






1 .68 


23 


62.4 




182.5 


2.24 






2.1 2 


24* 


66.0 




210 


2.72 






2.50 


25 


68.0 




245 


3.06 






2.86 


26* 


70.0 




280 


3.62 






3.20 


27 


71 .3 




301 . 5 


3.94 






3.52 


28 


72.0 




323.5 


4.22 






3.84 


29* 


73.5 




345 


4.84 






4.14 


30 


73.8 




412.5 


5.32 






4.44 


31* 


74.0 




480 


5.86 






4.74 


Aug 1 


77.5 




510 


6.14 






5.00 


2* 


81 .0 




540 


6.38 






5.23 


3 


82.0 




592.5 


6.73 






5.46 


4 


83.0 




645 


7.07 






5.68 


5 


84.0 




705 


7.63 






5.89 


6* 


85.0 




765 


8.25 






6.09 


7 


86.4 




787.5 


8.59 






6.29 


8* 


87.8 




81 


8.69 






6.48 


9 


90.2 




81 5 


9.35 






6.65 


10* 


92.8 




820 


9.55 






6.81 


11 


72.0 




808.5 


- 






7.14 


12 


51 . 




797 


10.1 6 






8.01 


13* 


30.0 




785 


10.34 






11 .56 


14 


25.5 




507.5 


10.64 






1 6.80 


1 5* 


21 .0 




125 


11.12 






24.66 


16 


27.0 




112.5 


11 .66 






28.96 


17* 


33.0 




100 


12.14 






31 .44 


18 


34.0 




11 


1 2.38 






33.77 


1 9* 


35.0 




1 20 


12.48 






35.96 


20 


40.5 




1 35 


1 2.58 






37.44 


Z = Water 


table de 


pth 


minus 40. 










h max = Soi 


1 moisture tension at a 


depth of 


4 cm. 





t 



A = Cumulative Class A pan evaporation. 

Cumulative daily rise calculated from properties of sample 5 

assuming 
*Dates of observation 

39. 7 mm 12 Aug. 



Z and h max are constant over each day. 



major rain events were 16 mm 10 Aug and 



- 36 - 



c 
o 

•r— 

oo 

c: 
<v 
-t-> 



-M 
oo 

c: 
o 
o 

CD 

■t-> 

oo 



X 

E 



CD 



CD 




-C 




S 




X 




<TJ 




E 




-C 




-o 




cz 




T3 




M 




4- 




O 




CO 




o> 




3 




i — 




A3 




> 


<D 




i — 


-o 


-Q 


o> 


ns 


+-> 


+-> 


CJ 




CD 


t_ 


1 — 


OJ 


CD 


-M 


CO 


03 




3 


+-> 




fC 


fT3 


ID 


OJ 




> 


CD 


O 


i — 


-Q 


<D_ 


OJ 


E 




<T3 


r-sl 


CO 






-M 


t_ 


_c 


O 


en 


4- 


• r— 




OJ 


>■> 


_c 


ro 




"O 


ro 



C 


"O 


c— 


CD 




CO 


X 


o 


rs 


CL 




E 






0) 



JD. 
"3 



O 
O 



o 



O 
00 



TO 

E 
o 

CD 

x to 

13 



o 

LO 



o 



o 

OO 



o 

(XI 






«3- 

O CM 
O O CM 

OOOCVJ'^HOCOWOWWCOfO 
OOOOOOCDO, — i — ■ — , — , — 



oo oo 



«3- 

o *d- 
o o ^t 

oooniooi 

CD O O O O O 



CO<3"lO NN N 



r-. 00 



to 

O VO 
CD CD tO 

oooiocri'tr^cnoro^<*5t 

CD CD CD CD O . — r— i— CM CJCMOJ OJ 



«* LO 
OJ OsJ 



o >— o 

oo. — cr»Lnojcocoo'=tuoLOLn in co 

CD CD CD CD . — CMOJOJOOOOOOOOOO oo oo 



OJ 

OWN 

OOi — inonoooincoro'vUmn m to 
oooi — ojoo^-^r«^-LOLf)LnLr> lolo 



O "vf «vl- 

ooroonfl'*'*mMcoOi — oj 
ooow^-ijDi^Nooooaicnai 



i — CO 

cri cn 



ONIO 
OONi — ^- 

ooo^ooifo^mirnoNrss 



OJ CD 

O OJ O 

CD CD OJ LO OJ CT> 



ro oo lo ld lo ld 
i — ojojrocococooooooo 



LD VO 

oo ro 



I--- CO 

o to co 

oo^anoai^ioNOiooo Oi — 

«d-cor-~cooocooocricr>cr> en cr> 



cm r-^ 

OCMr^-COCDLOtOOOJUDI — COCO OOCTi 



CM CTi to CTv CD . — i — i — i — i — i — 

i — cmcmcoooooooooooco 



oo oo 



CO 



X 



o o 
+ + 



O CD CD O O 
OCDCDOCDOOCDCDCD 
i — CSJ <d" CO 0O r- ICMOO^-LD 



+ + + 



+ + + + + 



^jhJr\ii^ji^j[^jr-ji\ir-jrvji^jr^jrM 



o o 
o o 
m o 



q 1 ncreases 
A2). 



- 37 - 



very rapidly as Z is decreased from 50 to 10 cm (Table 



Errors in measured h max , due to higher root densities near 
tensiometer tips, could lead to large errors in calculated rise 
when Z is small. For example, with Z = 20, true h max of 21 and 



measured h 



max 



of 100, the corresponding fluxes are 0.68 and 8.6 



cm/day respectively (Table A2) 



prope 

e q u i 1 

24-hr 

these 

20 wi 

(Tabl 

soil 

ri se 

occur 

24-hr 

s c i 1 
Z = 9 
At a 
flux 

1 day 
cm a n 
depen 
that , 
proje 
cons i 
and Z 
have 
Furth 
usage 
(and 

c o n d i 



Reca 
rt i e 
i bri 

per 

ste 
th p 
e Al 
moi s 
was 

du r 

per 
1 ati 
an 
suff 
i n e 

are 
das 
dene 

gi v 
ct i o 
de ra 

wi 1 
grad 
er, 

du r 
prob 
t i on 



11 th 
s usi 
urn wi 
i od a 
ady s 
a r t i c 
) and 
tu re 
1 ess 
i n g d 
iod ( 
on i n 
d day 
i ci en 
q u i 1 i 

appr 
s ume , 
e of 
en h m 
n i s 
t i on 
1 not 
ual ly 
at a 
ing d 
ably 

i s n 



at the cal 
ng Eq. 11, 
th an h max 
n d , with t 
t ate ass urn 
ul ar refer 

Table A2. 
tension gr 
than daily 
ayl i ght ho 
gi ven suff 

tensi on b 
t i me h = 5 
t ly 1 arge 
b r i u m with 
oxi mated. 

for irnmed 

q on n max 
ax = 500 
of course 
because st 
apply unt 
become re 
suf f i ci ent 
ayl i ght ho 
Z; Jaworsk 
ot likely 



c u 1 a t i o 
are va 
and Z 
his in 
pt i ons 
ence to 

As Z 

adual ly 

usage. 

urs and 

i ci ent 

ut no g 

00 (low 

value o 

a cons 

Now su 

i ate pu 

and Z a 

cm wate 

meaning 

eady st 

11 n max 
a d j u s t e 

ly smal 

urs, di 

i 1969) 

to be a 



ns of 
lido 
that 
mind, 
to tw 

the 
i n c r e 

i ncr 
Bee 

rise 
tensi 
reat 
er ri 
f Z t 
tant 
ppose 
rpose 
t sit 

p, fl 
1 ess 

ate r 

and 
d to 
1 val 
u rnal 
can 
pprox 



n se 

nly t 
remai 

cons 
o con 
f i rst 
ased 
eased 
ause 

woul 
on ) t 
di u rn 
ght h 
he as 
val ue 

Z is 
s , th 
e 2 e 
ux wi 
i n th 
e 1 a t i 
tensi 
the n 
ue of 

osci 
be ex 
i m a t e 



(Table 
o the e 
n const 
i d e r t h 
d i t i o n s 

run on 
from 51 

(Table 
most wa 
d occur 
here wi 
al osci 
and cor 
sumpt i o 

of h o 

sudden 
at Tabl 
xact ly . 
11 be 9 
e field 
o n s h i p s 
on : dept 
ew wate 

Zand 
1 1 at i on 
pected 
d. 



8), 

xtent 

ant o 

e rel 

; Z = 

site 

to 9 

Al), 

ter u 

th ro 

11 be 

1 lati 

ner o 

ns of 

vert 

ly de 

e A2 

Tab 
. cm 
cond 
betw 
h rel 
r tab 
with 
s i n 
and a 



from 

that 

ver e 

evanc 

90 a 

2 in 

3 cm, 

i.e. 

sage 

ughou 

some 

on i n 

f Tab 

cons 

he pe 

creas 

descr 

le A2 

/day. 

i t i o n 

een q 

a t i o n 

1 e de 

most 

n max 
stea 



i n 



core 

q is 
ach 
e of 
nd Z = 

1979 

dayt i me 

daily 
wou 1 d 
t the 

diurnal 

flux at 
le A2). 
tant 
riod of 
ed to 20 
i b e s the 

says 
Such a 
s under 

> h max 
ships 
pth. 
water 
and q 
dy state 



Estimates of rise, excluding these periods of shallow water 
table can be calculated as follows: 

Using the first run in 1979 as an example (Table Al), 
evapotranspiration for the period 20 July to 10 Aug can be 
calculated from class A pan evaporation records by assuming 
that the ratio of evapotranspiration to pan evaporation 
remains constant over the period of the run. Because the 
ground cover was well established before the start of the run 
(Table 1) and experienced only moderate moisture stress 
(Fig. 3) the error arising from this assumption is likely 
very small relative to the discrepancy under consideration. 



EV (ti) = [EV C (t 2 )/A (t 2 )] A (ti). 



- 38 - 



where ti = 20 July to 10 Aug, t2 = 20 July to 20 Aug, EV is 

evapot ranspi rat i on , A is class A pan evaporation and the 

subscripts o and c refer to open and closed cylinders 

respectively. 

From Tabl es 1 and Al , 

EV (ti ) = (8.35/12.58) x 9.55 

= 6. 34 cm. 

Soil moisture tensions on 10 Aug and 10 Sept were similar 
(Fig. 3). Therefore water contents (W ) will be similar, and 

W (10 Aug) = W (10 Sept) 

- 7.26 cm (Table 1 ). 

The accumulated deficit (D) for the period 20 July to 10 Aug 
wi 1 1 then be , 

D = w o 20 July - W 10 Aug 
= 11.80-7.26 

= 4.54 

and 

Rise (20 July to 10 Aug) = EV - D 

= 6.34-4.54 

= 1 . 80 cm. 

For this same period a rise of 6.81 cm was projected by 
sample 5. On this basis sample 5 is therefore in error by a 
factor of 6.81/1.8 = 3.8 (Table A3). A similar calculation 
for the 1977 run for the period 9 Aug to 16 Aug leads to an 
error factor of 4.0. The error factor for the second run on 
site 2 in 197 9 is 3.7 (Table A3). 



Therefore the combined effect of 1 or 2 (errors in 
measurement of K s and MRC and departure of sample 5 from profile 
properties at site #2) leads to estimates of rise that are 
approximately four times too large. Note however that this lack 
of agreement is a reflection of variability in measurements and is 
not a firm contradiction; the values of observed rise fell within 
the 95% confidence limits of calculated rise (Table A3). 

Tensiometer tips were positioned in disturbed soil (mean D^ 
of 1.45 g/cm 3 at 40-50 cm depth at site #2), 10 cm above the 
undisturbed subsoil (mean D ^ 1.83 g/cm 3 ), whereas a uniform 



39 - 



Table A3. Comparison of observed and calculated capillary rise 
at site #2 (sample 5) over periods free of unusually high water 
tabl e 



Peri od 



Observed 
ri se 
(cm) 



Cal cul ated 
rise 
(cm) 



Error 
(Calc./Obs. ) 



9 Aug-1 6 Aug, 1 977 

20 July-10 Aug, 1 979 

21 Aug-9 Sept, 1 979 



0.6 
1.8 
2.7 



2.4 (0.6-10.4) 

6.8 (1.6-29.6) 

10.1 (2.4-42.5) 



t 



4.0 
3.8 
3.7 



95% confidence 
cores . 



range based on rise calculated from each of four 



prof i 1 e 
the bel 
assumpt 
c o n d i t i 
than ob 
f a vou ra 
and a M 
of 1.47 
consequ 
combi na 
fa vou ra 
b (Eq. 
3.366 x 
con stan 
appreci 
disturb 



of 
ief 
i on 
on s . 
serv 
ble 
RC w 

g/c 
ence 
t i on 
ble 

9) I 
10 5 

ts o 

ably 
ed t 



und i 

that 
wou 1 
On 
ed r 
for 
e re 
m^ u 
s of 
s of 
for 
or t 

res 
f th 

lar 
h rou 



stu rbed 
the er 
d be sm 
e may s 
i se at 
c a p i 1 1 a 
measu re 
sing so 
this 1 
h and 
c a p i 1 1 a 
he di st 
pect i ve 
e u n d i s 
ger va 1 
ghout ( 



subso 
ror in 
all re 
pecul a 
site # 
ry ri s 
dona 
i 1 fro 
ay ered 
q , wer 
ry ri s 
u rbed 

ly, qu 

t u rbed 
ues of 

Table 



i 1 w 
t rod 

1 ati 
te t 

2 be 
e. 

cor 
in th 

s i t 
e i n 
e (T 
core 
ite 

cor 

Z i 
A4). 



as as 
uced 
ve to 
hat c 
cause 
To ex 
e tha 
is si 
u a t i o 

the 
able 

were 
di f fe 
es (T 
n pro 



sumed in the calculations in 
by this si mpl i fy i ng 

other errors under field 
alculated rise was larger 

the disturbed soil was less 
amine this possibility, K s 
t was reconstituted at a D^ 
t e . The i nd i cated 
n were small and, over 
opposite direcion, i.e 
A4) . The constants a , 

2. 359 x 10 5 , 4.568 and 
rent from the comparable 
able 5 , sampl e 5) and led 
files ass umed to be 



most 

more 
n and 



to 



However, replacing the upper 10 cm of undisturbed soil with 
disturbed soil had relatively little effect on Z in these 
simulations (Table A4). 



- 40 - 



Table A4. Height above a water table (Z) at which a tension of 
n max would lead to a flux of q for three assumed 
profile conditions at site #2 (sample 5) 



n max ( cm water) 



g = 1 cm/day 
500 100 20 



q = 0.1 c m / d a y 
500 100 20 



Z (cm) 



Assumed profile 

Sample 5, undisturbed 48.3 
Sample 5, disturbed 79.3 
10 cm sample 5 dist. over 
sample 5 undist. 53.5 



44.7 18.86 107.8 78.9 19.88 
72.9 19.79 132.0 93.9 19.98 

51 .7 19.59 105. 9 f 82.8 19.96 



t 



At n max = 500 cm and Q = 0*165 cm/day the undisturbed and layered 
profile simulations were equivalent, i.e. Z = 91 cm in 
undisturbed soil and Z = 91 cm in the layered profile composed of 
10 cm disturbed soil over 81 cm of undisturbed. At Z > 91 cm the 
layered profile is slightly less favourable to rise. 



- 41 - 



LIST OF FIGURES 



Fig. 1. Design of closed cylinder. 

Fig. 2. Top view of perforated concentric copper water dis- 
tribution tubes. 

Fig. 3. Soil moisture tension in centibars at a depth of 
40 cm (open cylinders) and depth to water table at sites 
1 and 2 over portions of three growing seasons. 

Fig. 4. Height (Z) above a stable water table and soil 
moisture tension (h) for selected values of steady state 
upward flux (q); sample no. 55. 



Page 
. 4 



13 



19 



- 42 - 



LIST OF TABLES 



Table 1. Schedule and particulars of capillary rise deter- 
minations at two sites 



Page 
5 



Table 2. Source of core samples 7 

Table 3. Moisture retention and other properties of core 14,15 
samples from 10 sites 



Table 4. Brooks and Corey constants (S r , h^ and X) and the 
fit of these constants to the moisture retention curve of 
core samples from 10 sites. 

Table 5. The constants a, n and b of Eqs. 9 and 10 for 
core samples from 10 sites 



Table 6. Constants a' and n' for core samples from 10 sites 
o values of h max where 
In q (cm/day) = a' + n'ln Z (cm) 



Table 7. Constants a" and n" for core samples from 10 
sites with h m a x = 4000 cm water where 

In Z (cm) = a" + n " 1 n q (mm/day) 
and values of Z at q = 10, 1 and 0.1 mm/day as obtained by 
numerical integration of Eq. 11 

Table 8. Comparison of observed capillary rise at two sites 
with rise calculated by numerical integration of Eq. 11 for 
the field conditions of daily Z (water table depth minus 40 
cm) and daily h max (soil moisture tension at 40 cm) 

Table 9. Relationships between hydraulic properties and 
other soil properties; correlation coefficients and 
regression equations, where 

In q (cm/day) = a' + n'ln Z (cm) and 

In Z (cm) = a" + n"ln q (mm/day) 

Table 10. Selected data and characteristics of the core 
samples arranged by type 

Table Al . Daily values of Z and h max at site 2 (1979), 
cumulative class A pan evaporation at the Kentville 
weather station and cumulative rise as calculated from 
sample 5 

Table A 2. Flux in cm/day for sample 5 at selected values 
of Z and h max where h max is the constant tension imposed 
at a height Z above a water table 



17 



20 



21 



22 



23 



26 



28,29 
35 



36 



- 43 - 



LIST OF TABLES (continued) 



Table A3. Comparison of observed and calculated capillary 
rise at site #2 (sample 5) over periods free of unusually 
hi gh water tabl e 



Table A4. 



Hei ght above a 
h, 



water 



table (Z) at which a 
tension of h max would lead to a flux of q for three 
assumed profile conditions at site #2 (sample 5) 



Page 
39 

40 



- 44 - 



h; 

0; 

0s ; 

S r ; 

r ; 

e s ; 

K; 

k s ; 
q; 

Z; 

Db; 

MRC; 



LIST OF COMMON SYMBOLS 

Soil moisture tension (cm water). 

Air entry value; see Eq. 5 (cm water). 

Soil water content (fraction of soil bulk volume; 
cm-^/cm^) . 

Soil water content measured at h = zero; i.e. not set 
equal to total porosity (cm^/cm^). 

Residualsaturation(=0 r /0 s ). 

Soil water content considered to be non -mobile; 

r = S r s . 

Effective saturation; see Eq. 4 (mobile water expressed 
as a fraction of unity). 

Hydraulic conductivity (cm/min). 

Saturated hydraulic conductivity (cm/min). 

Flux, positive upward (cm/day or mm/day). 

Height above a water table to some reference point (cm). 

Dry bulk density (g/cm^). 

Moisture retention curve; measured values of over a 
series of h values. 



a , n and b ; 
a ' and n ' ; 
a " and n " ; 
exp a" ; 



Constants of Eqs. 9 and 10. 

Con stants of Eq. 15 

Constants of Eq . 16 

From Eq. 16; the height Z (cm) over which a flux 
of 1 mm/day can be sustained by a constant tension 
of 4000 cm. 



LIBRARY / BIBUOTHEQUE 
AGRICULTURE CANADA OTTAWA K1A 0C5 

3 T073 ODDlEflDb fi 



630.72 Webster, D. H. 
C759 Measurement of capillary rise 

C 83- 1 9E under field conditions and related 

OOAg soil properties